Pith. sign in

REVIEW 4 major objections 6 minor 119 references

Privacy of Groups in Dense Street Imagery

T0 review · 4 major / 6 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Blurring faces and license plates in dense street imagery does not stop AI from inferring group membership, and authorities could exploit those inferences to target vulnerable groups such as street vendors and delivery workers.

desk verdict One solid empirical result (food trucks) carries the group-inference thesis; the delivery-worker experiment and the 'facial blurring' framing overreach, but the paper is still well worth a serious review. read the letter →

arxiv 2505.07085 v1 pith:RQOXI5F4 submitted 2025-05-11 cs.CY cs.CVcs.ET

Matt Franchi , Hauke Sandhaus , Madiha Zahrah Choksi , Severin Engelmann , Wendy Ju , Helen Nissenbaum This is my paper · ORCID
classification cs.CYcs.CVcs.ET
keywords privacydensestreetimagerygroupcontextualintegritycomputervisionsurveillancepenetrationtestingauditing
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper sets out to show that the standard privacy protection in dense street imagery — blurring faces and license plates before images are shared — does not protect the people pictured as members of groups. Working with 25,232,608 dashcam images of New York City in which pedestrians are already obscured, the authors run a penetration test: a vision-language model labels a sample of images, a lightweight detector is trained on those labels, and the detector maps food trucks and food delivery workers across the entire city in about 36 hours of compute on a single GPU. The food-truck map lands a median of 127 feet from recorded vending violations, and the delivery-worker map reproduces known hotspot patterns. The paper's point is that group membership, unlike individual identity, is inferable from context, attire, and equipment even when faces and bodies are blurred, and that this exposes already-targeted groups such as street vendors and delivery workers to enforcement. The authors then draw on contextual integrity to argue which uses of such inferred information are appropriate and which are not.

What carries the argument

The load-bearing mechanism is a cheap, fully automatable inference pipeline: zero-shot labeling by a vision-language model, human validation of a small subsample, training a lightweight YOLO object detector on those labels, and then running that detector over the entire dataset to produce (photograph, place, time) tuples for every instance of a group's visual signature. The paper supplements the pipeline with two conceptual tools. The first is a new de-identification failure mode, 'group membership inference': identically blurred objects of the same class can be clustered computationally, so the privacy protection itself becomes the signal an adversary clusters. The second is contextual integrity, which treats privacy as the appropriateness of information flow across five parameters — subject, sender, recipient, information type, and transmission principle — and turns the technical demonstration of inference into a normative argument about which flows are legitimate.

What would settle it

Station observers at the delivery-rider hotspots the zero-shot model flags during the 10AM-2PM lunch rush and count how many box-carrying riders are verifiable delivery couriers, judged by app insignia, restaurant dispatch, or self-identification; if confirmed couriers are not a large majority of flagged riders, the heatmap traces box-carrying cyclists rather than the vulnerable group. A companion calculation: compute the median distance from randomly placed citywide points to the nearest recorded vending violation; if that null median approaches the reported 127 feet, the food-truck detections add little locational signal over chance.

Watch

Extended reading notes

Core claim

The paper's central claim is that increased data density and advances in artificial intelligence enable harmful group membership inferences from supposedly anonymized street imagery, and that facial blurring provides no protection against an adversary analyzing group membership. The demonstration is a penetration test on a real-world dataset of 25,232,608 dashcam images collected in New York City: a zero-shot vision-language model labels roughly half a million images for the presence of food trucks, human annotators validate a subsample, a YOLO object detector trained on those labels processes the full dataset in 36 hours on a single GPU, and the resulting high-confidence detections lie a median of 127 feet from known food-truck vending violations. A second, fully zero-shot experiment asks the same model whether an image shows a bike rider with a box on their back, and converts the positive answers into delivery-worker hotspot maps for the lunch-rush period, with an estimated precision of 0.70. The paper argues these results generalize through a typology of identifiable groups — self-organized, role-based, cluster, and attribute-based — and names the underlying vulnerability 'group membership inference': when a provider blurs every instance of an object class in the same way, the blurred objects can be clustered to leak the group distribution the blurring was meant to conceal.

Load-bearing premise

The load-bearing assumption is that a bike rider with a storage box on their back is a food delivery worker: because the vision model could not reliably recognize delivery workers as such, the paper mapped the group by its equipment, and the delivery-worker hotspot analysis and the claim that blurring provides these workers no protection rest on that unvalidated visual proxy.

Editorial extensions

If this is right

  • Facial and body blurring is not a sufficient privacy guarantee for DSI datasets, because blurred rectangles of the same class can be clustered to reveal the group distribution the blurring was meant to hide.
  • An authority with DSI access can build a targeted-enforcement map of a vulnerable group in about 36 hours of compute on a single GPU, which the paper argues makes 'perfect enforcement' of vending rules a realistic prospect.
  • Researchers and providers should treat DSI sharing as a purpose-limited, risk-assessed flow governed by usage agreements and ethics oversight, since de-identification alone does not remove group-level harms.
  • The same inference pipeline points at every group in the paper's typology — nurses, protesters, religious communities, commuters — and the paper spells out the corresponding inappropriate flows and harms for each.
  • Anonymity and privacy are not the same thing: even when individuals cannot be identified, they can still be 'reachable' by authorities acting on group-level inferences, which is the paper's core normative conclusion.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The same zero-shot-to-detector pipeline could serve prosocial ends — public-health mapping of vending density, pedestrian planning, disaster response — so the paper's own framework implies the ethical status of the tool is set by recipient and transmission principle, not by the inference itself.
  • The 127-foot median is not compared against a null baseline; if randomly placed citywide points also sit within roughly a block of a known vending violation, part of the claimed locational signal could be an artifact of where violations concentrate.
  • As DSI archives accumulate over years, the single-season snapshots the pentest produces could be composed into longitudinal movement profiles of entire groups, an escalation the paper's density argument implies but does not demonstrate.
  • The purpose-limited sharing the paper recommends could be operationalized as technical access controls — query-level logging, purpose-bound API tiers, output filtering — rather than static data-use agreements, an implementation step the paper leaves open.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. The paper argues that dense street imagery (DSI), even after individual-level de-identification, enables harmful inferences about group membership. Using 25,232,608 Nexar dashcam images from New York City, the authors run two penetration-test experiments: (1) they use Cambrian-13B to zero-shot label food trucks, manually verify a sample, train a YOLOv11 detector, and deploy it across the full dataset, finding that high-confidence detections lie a median of 127 feet from known vending-violation records; (2) they use Cambrian-13B zero-shot to detect 'bike rider with a box on their back' as a proxy for food delivery workers, producing a hotspot heatmap. The paper then develops a typology of identifiable groups, applies contextual integrity to demarcate appropriate and inappropriate information flows, and offers policy and technical recommendations.

Significance. If the results hold, the paper makes an important empirical contribution to the privacy, accountability, and transparency literature: it demonstrates, on a real large-scale DSI dataset, that group-level inferences survive image obfuscation. The food truck experiment is the strongest part, because the detector's outputs are validated against independent public records (NYC OATH violation data) and the reported 127-foot median distance is a concrete, falsifiable measure. The paper also contributes a useful new de-identification failure mode, 'membership inference,' and a thoughtful contextual-integrity analysis of DSI information flows. The authors are transparent about several limitations, including the inability to verify the provider's sampling claims and the investigatory rather than production-grade nature of their models. However, the headline claim about facial blurring and food delivery workers is undercut by two load-bearing issues: the experiments use full-body obfuscation, not facial blurring, and the delivery-worker proxy is not validated against any external ground truth. These issues do not invalidate the broader group-inference thesis but do require revision.

major comments (4)
  1. [Appendix B.1.3] The annotation counts are internally inconsistent. The text states that of 2000 images sampled from Cambrian positives, 1496 contain food trucks and 645 do not, but 1496 + 645 = 2141, exceeding the quoted sample size. The reported TPR of 0.70 equals 1496/2141, whereas the correct precision on a 2000-image sample would be 1496/2000 = 0.748. Because this precision estimate characterizes the quality of the zero-shot labels used to train the YOLO model, the inconsistency affects the validity of the reported label quality and must be corrected (or the sampling procedure restated).
  2. [Section 5.1 and Introduction footnote] The central claim that 'facial blurring provides no protection for food delivery workers' is not directly supported by the experiments, because the Nexar dataset is full-body blurred rather than only facially blurred, as the introduction's footnote acknowledges. The experiments therefore demonstrate inference under full-body pedestrian obfuscation, not under facial-only blurring. A plausible monotonicity argument (full-body blurring removes more information than facial blurring, so if inference succeeds under full-body blurring it would also succeed under facial blurring) is needed to bridge this gap, but the paper does not state or justify it. Without this argument, the abstract's and Section 5.1's specific appeal to 'facial blurring' overstates what the experiments test.
  3. [Section 2.1, Experiment 2] The food delivery worker experiment relies on an unvalidated visual proxy: 'a bike rider with a box on their back.' The reported precision of 0.70 (from 500 random positive detections) validates only that the model's positives are bike riders with boxes, not that the riders are food delivery workers. The proxy can overcount other couriers (postal, freight, e-commerce) and undercount delivery workers using mopeds, cars, or different bag configurations. Consequently, the heatmap in Figure 3 is a map of box-carrying cyclists, and the Section 5.1 claim about food delivery workers specifically is not supported. In addition, treating Cambrian's zero-shot outputs as ground truth makes the heatmap largely a restatement of the VLM's own decisions; the paper should either validate the proxy with external ground truth or clearly reframe the claim as demonstrating inference of a visual proxy category.
  4. [Section 2.1.2 and Figure 2] The confidence-threshold reporting is ambiguous and undermines reproducibility of the headline result. Section 2.1.2 states that 'under an optimal confidence threshold of 0.205, the model asserted 196,183 images depicting food trucks,' while Figure 2 reports a confidence threshold of 0.7 with precision 0.90 and recall 0.50, and Section 2.1.3 refers to 'high-confidence' detections without specifying whether the 127-foot median uses the 0.7 threshold or another threshold. The paper should define 'optimal' explicitly and state which threshold is used for the vending-violation distance analysis.
minor comments (6)
  1. [Appendix B.1.2] The counts are inconsistent: the paper says 'approximately 500,000 randomly-sampled images' were queried, but then reports 2,903 positives and 557,602 negatives, which sum to 560,505; please clarify the actual number of images queried.
  2. [Ethical Considerations and Appendix B.1.1] The ethical statement that 'all data used in this study was collected during 2023' conflicts with the stated sampling period of August 11, 2023 to January 10, 2024 and with the note that data collection resumed on October 20, 2024 after an API overhaul; the statement should be corrected.
  3. [Section 2.1.1 and Ethical Considerations] The claim that the study 'deliberately avoid[s] cases where detection might lead to criminal consequences' is difficult to reconcile with the paper's own discussion of street vendors receiving over 1,200 criminal summonses and NYPD crackdowns on delivery workers' mopeds; please clarify or revise this assertion.
  4. [Figure 1 and Figure S1] Figure 1's caption describes the data as 'facially de-identified,' while Figure S1 shows that the provider blurs entire pedestrian bodies; the terminology should be harmonized throughout the paper.
  5. [Section 5.3.1] There is a typo: 'downstream downstream efforts' should read 'downstream efforts.'
  6. [Reference [56]] Reference [56] has an unmatched parenthesis after '3 trillion images'; please correct the citation.

Circularity Check

1 steps flagged · score 3.0 of 10

Partial circularity in the delivery-worker experiment: the heatmap relabels Cambrian's 'bike rider with a box' detections as 'food delivery workers'; the food truck result is externally validated and non-circular.

  1. renaming known result [Section 2.1 Experiment 2, Section 2.1.3, Section 5.1, Figure 3 caption]
    "For this task, we asked Cambrian: 'Is there a bike rider with a box on their back in this image?' Then, we took the Cambrian model output as ground-truth and created a detection heatmap ... From the detections of food delivery workers in Experiment 2, we are also able to easily create targeted deployment zones for the in-the-wild surveillance of food delivery workers."

    The VLM was not asked to detect the target group, 'food delivery worker'; the authors state in a footnote that Cambrian 'has little predictive power on domain-specific terms like food delivery worker' and instead prompted for a visual proxy, 'food storage boxes strapped onto the back of a bike.' The heatmap is therefore a map of box-carrying cyclists as judged by Cambrian, yet Section 2.1.3 and Figure 3 relabel these detections as 'food delivery workers.' The paper's own precision estimate of 0.70 validates only that positive detections contain a bike rider with a box, not that the rider belongs to the vulnerable group.

full rationale

The food truck penetration test is self-contained and non-circular: Cambrian's zero-shot positives were human-validated, a YOLO model was trained on those labels, and the resulting high-confidence detections were compared against independent NYC OATH vending-violation records; the median distance of 127 feet is an external benchmark rather than a fitted output. The contextual integrity analysis and recommendations are normative framing and do not claim a derivation. The one partially circular element is Experiment 2's delivery-worker result. Because Cambrian could not detect 'food delivery worker' directly, the authors chose the proxy 'bike rider with a box,' treated Cambrian's output as ground truth for a heatmap, and then in Sections 2.1.3 and 5.1 described these detections as 'food delivery workers' and concluded that 'facial blurring provides no protection for food delivery workers.' The heatmap is a relabeled rendering of the proxy detections, not an independent inference about group membership; the reported precision validates only the visual proxy. This is a renaming/operationalization circularity limited to the delivery-worker component of the central claim. A further non-circular validity caveat is that the dataset is full-body blurred rather than only facially blurred, so the Section 5.1 statement about facial blurring is an extrapolation beyond the tested obfuscation. Overall, the main empirical result is externally grounded, but the delivery-worker claim partially reduces to its input, supporting a score of 3.

Assumptions & free parameters 2 free parameters · 3 assumptions · 0 invented entities

The demonstration rests on three unverified premises: the Nexar dataset is representative of NYC street imagery (the authors explicitly flag this in B.1.1), a bike rider with a box is a valid proxy for a food delivery worker, and Cambrian's zero-shot outputs can serve as ground truth in Experiment 2. The deployment confidence threshold is hand-chosen and affects the reported food truck counts.

free parameters (2)
  • Deployment confidence threshold for food truck detections = 0.205 (high-precision map uses 0.7)
    The paper reports 196,183 food truck images 'under an optimal confidence threshold of 0.205' but does not define the optimality criterion; the count and maps depend on this hand-chosen threshold.
  • High-precision threshold for Figure 2 = 0.7
    Chosen to yield precision 0.90 and recall 0.50 on the test set; the overlap analysis with vending violations uses this threshold, so the median-distance finding depends on it.
assumptions (3)
  • domain assumption The Nexar dataset is a representative sample of NYC street imagery from the camera network.
    The authors state in B.1.1 they cannot independently verify random sampling or access camera network statistics; claims about city-wide group distributions depend on this unverified representativeness.
  • ad hoc to paper A bike rider with a box strapped to their back is a valid proxy for a food delivery worker.
    Experiment 2 prompts Cambrian for this proxy because the model failed on the direct term 'food delivery worker'; the delivery-worker heatmap inherits this assumption, which is not validated against true occupation labels.
  • domain assumption Cambrian zero-shot outputs can serve as ground truth for training and evaluation in Experiment 2.
    For delivery riders, the paper uses Cambrian's own outputs as ground truth and estimates precision on a 500-image sample; the heatmap and hotspot analysis rest on this assumption.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Privacy of Groups in Dense Street Imagery." pith.science (2026). https://pith.science/paper/RQOXI5F4

@misc{pith2026250507085,
  author       = {Pith},
  title        = {Pith review of: Privacy of Groups in Dense Street Imagery},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RQOXI5F4}},
  note         = {Machine review of arXiv:2505.07085}
}
read the original abstract

Spatially and temporally dense street imagery (DSI) datasets have grown unbounded. In 2024, individual companies possessed around 3 trillion unique images of public streets. DSI data streams are only set to grow as companies like Lyft and Waymo use DSI to train autonomous vehicle algorithms and analyze collisions. Academic researchers leverage DSI to explore novel approaches to urban analysis. Despite good-faith efforts by DSI providers to protect individual privacy through blurring faces and license plates, these measures fail to address broader privacy concerns. In this work, we find that increased data density and advancements in artificial intelligence enable harmful group membership inferences from supposedly anonymized data. We perform a penetration test to demonstrate how easily sensitive group affiliations can be inferred from obfuscated pedestrians in 25,232,608 dashcam images taken in New York City. We develop a typology of identifiable groups within DSI and analyze privacy implications through the lens of contextual integrity. Finally, we discuss actionable recommendations for researchers working with data from DSI providers.

Figures

Figures reproduced from arXiv: 2505.07085 by the authors.

Figure 1
Figure 1. AI-inferred group membership in a dataset of more than 25 million facially de-identified dashcam images from NYC [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. A map showing reported vending violations against [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Using zero-shot image retrieval, we queried Cam [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Contextual Integrity Analysis of a DSI Information Flow. Changing a single parameter in an information flow can [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

119 extracted references · 54 canonical work pages

  1. [1]

    [n. d.]. Blur or remove 360 imagery & Photo Paths - Computer - Google Maps Help. https://support.google.com/maps/answer/7011973?hl=en&co=GENIE. Platform%3DDesktop

    work page arXiv

  2. [2]

    New Report Shows Mayor Adams, Commissioner Mayuga Deliver for Delivery Workers by Significantly Boos

    2024. New Report Shows Mayor Adams, Commissioner Mayuga Deliver for Delivery Workers by Significantly Boos. http://www.nyc.gov/office-of- the-mayor/news/539-24/new-report-shows-mayor-adams-commissioner- mayuga-deliver-delivery-workers-significantly

    2024

  3. [3]

    Vendors rally against NYPD crackdown, call for more licensing in New York City

    2024. Vendors rally against NYPD crackdown, call for more licensing in New York City. https://abc7ny.com/street-vendors-rally-against-nypd-crackdown- set-to-hold-march-call-for-more-licensing-in-new-york-city/14687737/ Sec- tion: crime-safety

    work page arXiv 2024

  4. [4]

    Herman Aguinis and Kyle J Bradley. 2014. Best practice recommendations for designing and implementing experimental vignette methodology studies. Organizational research methods 17, 4 (25 oct 2014), 351–371. https://doi.org/10. 1177/1094428114547952

    2014

  5. [5]

    Tim Alpherts, Sennay Ghebreab, Yen-Chia Hsu, and Nanne Van Noord. 2024. Perceptive Visual Urban Analytics is Not (Yet) Suitable for Municipalities. In The 2024 ACM Conference on Fairness, Accountability, and Transparency . ACM, Rio de Janeiro Brazil, 1341–1354. https://doi.org/10.1145/3630106.3658976

    arXiv 2024

  6. [6]

    Dragomir Anguelov, Carole Dulong, Daniel Filip, Christian Frueh, Stéphane Lafon, Richard Lyon, Abhijit Ogale, Luc Vincent, and Josh Weaver. 2010. Google Street View: Capturing the World at Street Level. Computer 43, 6 (June 2010), 32–38. https://doi.org/10.1109/MC.2010.170 Conference Name: Computer

    doi:10.1109/mc.2010.170 2010

  7. [7]

    Haleh Asgarinia. 2024. Limiting access to certain anonymous information: From the group right to privacy to the principle of protecting the vulnerable. The Journal of value inquiry (23 apr 2024), 1–27. https://doi.org/10.1007/s10790- 024-09980-x

    work page doi:10.1007/s10790- 2024

  8. [8]

    Dewan Mehrab Ashrafi. 2024. Technology Takes the Wheel: Unveiling the Drivers of Car Dashcam Adoption. International Journal of Innovation and Technology Management 21, 03 (May 2024), 2450024. https://doi.org/10.1142/ S021987702450024X Publisher: World Scientific Publishing Co

    2024

Show all 119 references
  1. [9]

    Cedar Attanasio. 2024. Getting Paid Now More Complex for NYC Food Delivery Workers. https://www.ttnews.com/articles/new-york-food-delivery-workers

    2024

  2. [10]

    Michael D. M. Bader, Stephen J. Mooney, Blake Bennett, and Andrew G. Rundle

  3. [11]

    Michael Bailey, David Dittrich, Erin Kenneally, and Doug Maughan. 2012. The Menlo Report. IEEE Security & Privacy 10, 2 (March 2012), 71–75. https: //doi.org/10.1109/MSP.2012.52 Conference Name: IEEE Security & Privacy

    2012 doi

  4. [12]

    Solon Barocas and Helen Nissenbaum. 2014. Big Data’s End Run around Anonymity and Consent. In Privacy, Big Data, and the Public Good: Frame- works for Engagement, Helen Nissenbaum, Julia Lane, Stefan Bender, and Vic- toria Stodden (Eds.). Cambridge University Press, Cambridge,...

    2014 doi

  5. [13]

    Mitchell, and Helen Nissenbaum

    Adam Barth, Anupam Datta, John C. Mitchell, and Helen Nissenbaum. 2006. Privacy and contextual integrity: Framework and applications. In 2006 IEEE symposium on security and privacy (S&P’06) . IEEE, 15–pp. https://ieeexplore. ieee.org/abstract/document/1624011/

    2006

  6. [14]

    Lyria Bennett Moses and Janet Chan. 2018. Algorithmic prediction in policing: assumptions, evaluation, and accountability. Policing and Society 28, 7 (Sept. 2018), 806–822. https://doi.org/10.1080/10439463.2016.1253695

    2018

  7. [15]

    Dulari Bhatt, Chirag Patel, Hardik Talsania, Jigar Patel, Rasmika Vaghela, Sharnil Pandya, Kirit Modi, and Hemant Ghayvat. 2021. CNN variants for computer vision: History, architecture, application, challenges and future scope.Electronics 10, 20 (2021), 2470

    2021

  8. [16]

    Matt Bishop. 2007. About Penetration Testing. IEEE Security & Privacy 5, 6 (Nov. 2007), 84–87. https://doi.org/10.1109/MSP.2007.159 Conference Name: IEEE Security & Privacy

    2007 doi

  9. [17]

    Bloustein and Nathaniel J

    Edward J. Bloustein and Nathaniel J. Pallone. 2017.Individual and Group Privacy. Routledge, New York. https://doi.org/10.4324/9781351319966

    2017 doi

  10. [18]

    Lukas Bossard, Matthias Dantone, Christian Leistner, Christian Wengert, Till Quack, and Luc Van Gool. 2013. Apparel classification with style. In Computer Vision–ACCV 2012: 11th Asian Conference on Computer Vision, Daejeon, Korea, November 5-9, 2012, Revised Selected Papers, P...

    2013

  11. [19]

    August Bourgeus, Laurens Vandercruysse, and Nanouk Verhulst. 2024. Under- standing contextual expectations for sharing wearables’ data: Insights from a vignette study. Computers in human behavior reports 15, 100443 (1 aug 2024), 100443. https://doi.org/10.1016/j.chbr.2024.100443

    2024

  12. [20]

    Mark Burdon, Tegan Cohen, Josh Buckley, and Michael Milford. 2024. From object obfuscation to contextually-dependent identification: enhancing au- tomated privacy protection in street-level image platforms (SLIPs). In- formation & Communications Technology Law 33, 2 (May 2024)...

    2024

  13. [21]

    Mark Burdon and Alissa McKillop. 2014. The Google street view Wi-Fi scandal and its repercussions for privacy regulation. Monash University Law Review 39, 3 (Jan. 2014), 702–738. https://doi.org/10.3316/informit.376209506308923 Publisher: Monash University - Faculty of Busines...

    2014 doi

  14. [22]

    Richard Campanella. 2017. People-Mapping Through Google Street View.Places Journal (Nov. 2017). https://doi.org/10.22269/171114

    2017 doi

  15. [23]

    Filippo Cavallo, Francesco Semeraro, Laura Fiorini, Gergely Magyar, Peter Sinčák, and Paolo Dario. 2018. Emotion modelling for social robotics applica- tions: a review. Journal of Bionic Engineering 15 (2018), 185–203

    2018

  16. [24]

    Urban Justice Center. 2023. Street Vendor Project. https://www.streetvendor. org/what-is-svp

    2023

  17. [25]

    David L. Chaum. 1981. Untraceable electronic mail, return addresses, and digital pseudonyms. Commun. ACM 24, 2 (Feb. 1981), 84–90. https://doi.org/10.1145/ 358549.358563

    1981

  18. [26]

    Wen-Huang Cheng, Sijie Song, Chieh-Yun Chen, Shintami Chusnul Hidayati, and Jiaying Liu. 2021. Fashion meets computer vision: A survey.ACM Computing Surveys (CSUR) 54, 4 (2021), 1–41

    2021

  19. [27]

    Madiha Zahrah Choksi, Ero Balso, Frauke Kreuter, and Helen Nissenbaum

  20. [28]

    Haidee Chu. 2024. NYPD Dragging Many More Vendors to Criminal Court, Data Shows. http://www.thecity.nyc/2024/02/05/nypd-vendors-criminal- summonses-court-spike/

    2024

  21. [29]

    Chris Clews, Roza Brajkovich-Payne, Emily Dwight, Ayob Ahmad Fauzul, Madeleine Burton, Olivia Carleton, Julie Cook, Charlotte Deroles, Ruby Faulkner, Mary Furniss, Anahera Herewini, Daymen Huband, Nerissa Jones, Cho Wool Kim, Alice Li, Jacky Lu, James Stanley, Nick Wilson, and...

  22. [30]

    Etienne Corvee, Slawomir Bak, and François Bremond. 2012. People detection and re-identification for multi surveillance cameras. https://inria.hal.science/ hal-00656108

    2012

  23. [31]

    Bahar Dadashova, Chiara Silvestri Dobrovolny, Mahmood Tabesh, Safety through Disruption (Safe-D) University Transportation Center (UTC), and Texas A&M Transportation Institute. 2021. Detecting Pavement Distresses Using Crowd- sourced Dashcam Camera Images . Technical Report TT...

    2021

  24. [32]

    Hidalgo, Michel Verleysen, and Vin- cent D

    Yves-Alexandre De Montjoye, César A. Hidalgo, Michel Verleysen, and Vin- cent D. Blondel. 2013. Unique in the crowd: The privacy bounds of human mobility. Scientific reports 3, 1 (2013), 1–5. https://doi.org/10.1038/srep01376 Publisher: Nature Publishing Group

    2013 doi

  25. [33]

    Yves-Alexandre De Montjoye, Laura Radaelli, Vivek Kumar Singh, and Alex “Sandy” Pentland. 2015. Unique in the shopping mall: On the reiden- tifiability of credit card metadata. Science 347, 6221 (Jan. 2015), 536–539. https://doi.org/10.1126/science.1256297

    2015 doi

  26. [34]

    Dietrich and Melissa L

    Bryce J. Dietrich and Melissa L. Sands. 2023. Seeing racial avoidance on New York City streets. Nature Human Behaviour 7, 8 (Aug. 2023), 1275–1281. https: //doi.org/10.1038/s41562-023-01589-7 Publisher: Nature Publishing Group

    2023 doi

  27. [35]

    Sarah Elwood and Agnieszka Leszczynski. 2011. Privacy, reconsidered: New representations, data practices, and the geoweb. Geoforum 42, 1 (Jan. 2011), 6–15. https://doi.org/10.1016/j.geoforum.2010.08.003

    2011 doi

  28. [36]

    Severin Engelmann, Madiha Zahrah Choksi, Angelina Wang, and Casey Fiesler

  29. [37]

    Severin Engelmann and Helen Nissenbaum. 2025. Countering Privacy Nihilism. In Conceptions of Data Protection and Privacy: Legal and Philosophical Perspectives, Elisa Orrù and Ralf Poscher (Eds.). Hart Publishing

    2025

  30. [38]

    Severin Engelmann, Chiara Ullstein, Orestis Papakyriakopoulos, and Jens Grossklags. 2022. What people think AI should infer from faces. In Proceed- ings of the 2022 ACM Conference on Fairness, Accountability, and Transparency . 128–141

    2022

  31. [39]

    Arturo Flores and Serge Belongie. 2010. Removing pedestrians from Google street view images. In2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Workshops. 53–58. https://doi.org/10.1109/CVPRW. 2010.5543255 ISSN: 2160-7516

    2010

  32. [40]

    In Proceedings of the 2024 ACM Conference on Fairness, Accountability, and Transparency

    Visions of a discipline: Analyzing introductory AI courses on YouTube. In Proceedings of the 2024 ACM Conference on Fairness, Accountability, and Transparency. 2400–2420

    2024

  33. [41]

    Jessica Formoso. 2024. Bronx bodega delivery workers targeted in string of robberies. https://www.fox5ny.com/news/bronx-bodega-delivery-workers- targeted-string-robberies Publisher: FOX 5 New York

    2024

  34. [42]

    Matt Franchi, Debargha Dey, and Wendy Ju. 2024. Towards Instrumented Fingerprinting of Urban Traffic: A Novel Methodology using Distributed Mobile Point-of-View Cameras. In Proceedings of the 16th International Conference on Automotive User Interfaces and Interactive Vehicular...

    2024

  35. [43]

    Matt Franchi, Nikhil Garg, Wendy Ju, and Emma Pierson. 2025. Bayesian Modeling of Zero-Shot Classifications for Urban Flood Detection. https: //doi.org/10.48550/arXiv.2503.14754 arXiv:2503.14754 [cs]

    2025 doi arXiv

  36. [44]

    Luciano Floridi. 2017. Group Privacy - A Defense and an Interpretation. https: //doi.org/10.2139/ssrn.3854483

    2017 doi

  37. [45]

    Zamfirescu-Pereira, Wendy Ju, and Emma Pierson

    Matt Franchi, J.D. Zamfirescu-Pereira, Wendy Ju, and Emma Pierson. 2023. Detecting disparities in police deployments using dashcam data. In Proceedings of the 2023 ACM Conference on Fairness, Accountability, and Transparency (FAccT ’23). Association for Computing Machinery, Ne...

    2023

  38. [46]

    Andrea Frome, German Cheung, Ahmad Abdulkader, Marco Zennaro, Bo Wu, Alessandro Bissacco, Hartwig Adam, Hartmut Neven, and Luc Vincent. 2009. Large-scale privacy protection in Google Street View. In 2009 IEEE 12th Interna- tional Conference on Computer Vision . 2373–2380. http...

    2009

  39. [47]

    Patrick Gallo and Houssain Kettani. 2020. On Privacy Issues with Google Street View CLEAR Conference Computer Science Academic Papers.South Dakota Law Review 65, 3 (2020), 608–622. https://heinonline.org/HOL/P?h=hein.journals/ sdlr65&i=666

    2020

  40. [48]

    Matthew Franchi, Maria Teresa Parreira, Fanjun Bu, and Wendy Ju. 2025. The Robotability Score: Enabling Harmonious Robot Navigation on Urban Streets. In Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems . ACM, Yokohama Japan, 1–17. https://doi.org/10...

    2025

  41. [49]

    R Stuart Geiger, Kevin Yu, Yanlai Yang, Mindy Dai, Jie Qiu, Rebekah Tang, and Jenny Huang. 2020. Garbage in, garbage out? Do machine learning application papers in social computing report where human-labeled training data comes from?. In Proceedings of the 2020 conference on f...

    2020

  42. [50]

    Roger C Geissler. [n. d.]. Private Eyes Watching You: Google Street View and the Right to an Inviolate Personality. HASTINGS LA W JOURNAL63 ([n. d.])

  43. [51]

    Jake Goldenfein. 2019. The profiling potential of computer vision and the challenge of computational empiricism. In Proceedings of the Conference on Fairness, Accountability, and Transparency. 110–119

    2019

  44. [52]

    Timnit Gebru, Jonathan Krause, Yilun Wang, Duyun Chen, Jia Deng, Erez Lieber- man Aiden, and Li Fei-Fei. 2017. Using deep learning and Google Street View to estimate the demographic makeup of neighborhoods across the United States. Proceedings of the National Academy of Scienc...

    2017 doi

  45. [53]

    Ben Green. 2019. The Smart Enough City: Putting Technology in Its Place to Reclaim Our Urban Future . MIT Press. Google-Books-ID: avGRDwAAQBAJ

    2019

  46. [54]

    Tim Gruchmann and Amer Jazairy. 2025. Big brother is watching you: Examining truck drivers’ acceptance of road-facing dashcams. Transportation Research Part F: Traffic Psychology and Behaviour 111 (May 2025), 316–330. https://doi.org/ 10.1016/j.trf.2025.03.015

    2025 doi

  47. [55]

    Aya Hassouneh, AM Mutawa, and M Murugappan. 2020. Development of a real-time emotion recognition system using facial expressions and EEG based on machine learning and deep neural network methods. Informatics in Medicine Unlocked 20 (2020), 100372

    2020

  48. [56]

    Google. 2025. Google-Contributed Street View Imagery Policy. https://www. google.com/streetview/policy/

    2025

  49. [57]

    Mark Healy. 2024. How the NYPD’s Scooter Crackdown Beat Down Food Delivery Workers. https://www.curbed.com/article/moped-crackdown-nypd- seizure-delivery-workers-erie-basin.html

    2024

  50. [58]

    Marco Helbich, Matthew Danish, S. M. Labib, and Britta Ricker. 2024. To use or not to use proprietary street view images in (health and place) research? That is the question. Health & Place 87 (May 2024), 103244. https://doi.org/10.1016/j. healthplace.2024.103244

    2024

  51. [59]

    Sam Hind and Alex Gekker. 2024. Automotive parasitism: Examining Mobileye’s ‘car-agnostic’ platformisation. New Media & Society 26, 7 (July 2024), 3707–3727. https://doi.org/10.1177/14614448221104209 Publisher: SAGE Publications

    2024 doi

  52. [60]

    Andrew J. Hawkins. 2024. Lyft is also partnering with robotaxi companies. The Verge (Nov. 2024). https://www.theverge.com/2024/11/6/24289475/lyft-may- mobility-mobileye-nexar-autonomous-robotaxi

    2024

  53. [61]

    Joel Janai, Fatma Güney, Aseem Behl, Andreas Geiger, et al. 2020. Computer vision for autonomous vehicles: Problems, datasets and state of the art. Founda- tions and Trends® in Computer Graphics and Vision 12, 1–3 (2020), 1–308

    2020

  54. [62]

    Glenn Jocher, Jing Qiu, and Ayush Chaurasia. 2023. Ultralytics YOLO. https: //github.com/ultralytics/ultralytics original-date: 2022-09-11T16:39:45Z

    2023

  55. [63]

    Dimitrios Kastaniotis, Ilias Theodorakopoulos, George Economou, and Spiros Fotopoulos. 2013. Gait-based gender recognition using pose information for real time applications. In 2013 18th international conference on digital signal processing (DSP). IEEE, 1–6

    2013

  56. [64]

    Lazar Ilic, Michael Sawada, and Amaury Zarzelli. 2019. Deep mapping gentri- fication in a large Canadian city using deep learning and Google Street View. PloS one 14, 3 (2019), e0212814. https://journals.plos.org/plosone/article?id=10. 1371/journal.pone.0212814 Publisher: Publ...

    2019

  57. [65]

    Kanako Kawaguchi and Yukiko Kawaguchi. 2012. What Does Google Street View Bring about? -Privacy, Discomfort and The Problem of Paradoxical Others- . Contemporary and Applied Philosophy 4 (Aug. 2012), 19–34. https://doi.org/ 10.14989/180276 Accepted: 2014-01-15T04:33:11Z

    2012 doi

  58. [66]

    Junghwan Kim and Kee Moon Jang. 2023. An examination of the spatial coverage and temporal variability of Google Street View (GSV) images in small- and medium-sized cities: A people-based approach. Computers, Environment and Urban Systems 102 (June 2023), 101956. https://doi.or...

    2023

  59. [67]

    Anil Kunchala, Melanie Bouroche, and Bianca Schoen-Phelan. 2023. Towards A Framework for Privacy-Preserving Pedestrian Analysis. In 2023 IEEE/CVF Winter Conference on Applications of Computer Vision (W ACV). IEEE, Waikoloa, HI, USA, 4359–4369. https://doi.org/10.1109/WACV56688...

    2023

  60. [68]

    Fareed Kaviani, Ben Lyall, and Sjaan Koppel. 2024. Exploring social perceptions of everyday smartglass use in Australia. PLOS ONE 19, 11 (Nov. 2024), e0313001. https://doi.org/10.1371/journal.pone.0313001 Publisher: Public Library of Sci- ence

    2024 doi

  61. [69]

    Michael J. V. Leach, Rolf Baxter, Neil M. Robertson, and Ed P. Sparks. 2014. Detect- ing Social Groups in Crowded Surveillance Videos Using Visual Attention. 461–

    2014

  62. [70]

    Michele Loi and Markus Christen. 2020. Two concepts of group privacy. Philos- ophy & technology 33, 2 (June 2020), 207–224. https://doi.org/10.1007/s13347- 019-00351-0

    2020 doi

  63. [71]

    Kristian Lum and William Isaac. 2016. To predict and serve? Significance 13, 5 (2016), 14–19. https://doi.org/10.1111/j.1740-9713.2016.00960.x Publisher: Oxford University Press

    2016

  64. [72]

    Julia Lane, Victoria Stodden, Stefan Bender, and Helen Nissenbaum. 2014. Pri- vacy, big data, and the public good: Frameworks for engagement . Cambridge University Press, Cambridge, England

    2014

  65. [73]

    Alessandro Mantelero. 2017. From group privacy to collective privacy: Towards a new dimension of privacy and data protection in the big data era. In Group Privacy. Springer International Publishing, Cham, 139–158. https://doi.org/10. 1007/978-3-319-46608-8_8

    2017

  66. [74]

    Coral Murphy Marcos. 2021. As Bike Thefts Jump, Delivery Workers Band Together for Safety. The New York Times (Oct. 2021). https://www.nytimes. com/2021/10/12/business/delivery-workers-thefts-neighborhood-watch.html

    2021

  67. [75]

    Steve Matthews. 2010. Anonymity and the Social Self. American Philosoph- ical Quarterly 47, 4 (2010), 351–363. https://www.jstor.org/stable/25734161 Publisher: [North American Philosophical Publications, University of Illinois Press]

    2010

  68. [76]

    Lorraine Mazerolle, David Hurley, and Mitchell Chamlin. 2002. Social Behavior in Public Space: An Analysis of Behavioral Adaptations to CCTV. Security Journal 15, 3 (July 2002), 59–75. https://doi.org/10.1057/palgrave.sj.8340118

    2002 doi

  69. [77]

    Kevin Macnish. 2012. Unblinking eyes: the ethics of automating surveillance. Ethics and information technology 14, 2 (7 jun 2012), 151–167. https://doi.org/ 10.1007/s10676-012-9291-0

    2012 doi

  70. [78]

    Nikhil Naik, Jade Philipoom, Ramesh Raskar, and César Hidalgo. 2014. Streetscore-predicting the perceived safety of one million streetscapes. In Proceedings of the IEEE conference on computer vision and pat- tern recognition workshops . 779–785. https://www.cv-foundation.org/ ...

    2014

  71. [79]

    Nguyen, Sahil Khanna, Pallavi Dwivedi, Dina Huang, Yuru Huang, Tolga Tasdizen, Kimberly D

    Quynh C. Nguyen, Sahil Khanna, Pallavi Dwivedi, Dina Huang, Yuru Huang, Tolga Tasdizen, Kimberly D. Brunisholz, Feifei Li, Wyatt Gorman, and Thu T. Nguyen. 2019. Using Google Street View to examine associations between built environment characteristics and US health outcomes. ...

    2019

  72. [80]

    Helen Nissenbaum. 2004. Privacy as contextual integrity. Wash. L. Rev. 79 (2004), 119. https://heinonline.org/hol-cgi-bin/get_pdf.cgi?handle=hein. journals/washlr79&section=16 Publisher: HeinOnline

    2004

  73. [81]

    Helen Nissenbaum. 2011. A Contextual Approach to Privacy Online. Daedalus 140, 4 (Oct. 2011), 32–48. https://doi.org/10.1162/DAED_a_00113

    2011 doi

  74. [82]

    Mobileye. 2022. Mobileye’s Self-Driving Secret? 200PB of Data | Mobileye Blog. https://www.mobileye.com/blog/mobileye-ces-2022-self-driving-secret-data/

    2022

  75. [83]

    Angelo Nodari, Marco Vanetti, and Ignazio Gallo. 2012. Digital privacy: Replacing pedestrians from Google Street View images. In Proceedings of the 21st International Conference on Pattern Recognition (ICPR2012) . 2889–2893. https://ieeexplore.ieee.org/abstract/document/646076...

    2012

  76. [84]

    nycadmin. 2014. New York City Food by the Numbers: Food Trucks. https: //www.nycfoodpolicy.org/new-york-city-food-numbers-food-trucks/

    2014

  77. [85]

    NYCOpenData. 2025. OATH Hearings Division Case Status. https: //data.cityofnewyork.us/City-Government/OATH-Hearings-Division-Case- Status/jz4z-kudi FAccT ’25, June 23–26, 2025, Athens, Greece Franchi and Sandhaus et al

    2025

  78. [86]

    Ojeda, Ambar Reyes, April Xu, CEDAR ATTANASIO Documented, and Associated Press• •

    Rommel H. Ojeda, Ambar Reyes, April Xu, CEDAR ATTANASIO Documented, and Associated Press• •. 2024. Newly arrived migrants encounter hazards of food delivery on NYC streets: robbers. https://www.nbcnewyork.com/news/local/ safety-migrants-encounter-hazards-food-delivery-nyc-robb...

    2024

  79. [87]

    Helen Nissenbaum. 2019. Contextual Integrity Up and Down the Data Food Chain. Theoretical Inquiries in Law 20, 1 (Jan. 2019), 221–256. https://doi.org/ 10.1515/til-2019-0008 Publisher: De Gruyter

    2019 doi

  80. [88]

    Eugenia Politou, Efthimios Alepis, and Constantinos Patsakis. 2018. Forgetting personal data and revoking consent under the GDPR: Challenges and proposed solutions. Journal of Cybersecurity 4, 1 (Jan. 2018). https://doi.org/10.1093/ cybsec/tyy001

    2018

  81. [89]

    Apostolos Pyrgelis, Carmela Troncoso, and Emiliano De Cristofaro. 2017. Knock knock, who’s there? Membership inference on aggregate location data. arXiv preprint arXiv:1708.06145 (2017)

    2017 arXiv

  82. [90]

    Redmill, Ekim Yurtsever, Rabi G

    Keith A. Redmill, Ekim Yurtsever, Rabi G. Mishalani, Benjamin Coifman, and Mark R. McCord. 2023. Automated Traffic Surveillance Using Existing Cameras on Transit Buses. Sensors 23, 11 (Jan. 2023), 5086. https://doi.org/10.3390/ s23115086 Number: 11 Publisher: Multidisciplinary...

    2023

  83. [91]

    Kate Robertson, Cynthia Khoo, and Yolanda Song. 2020. To surveil and predict: A human rights analysis of algorithmic policing in Canada. (2020). https: //citizenlab.ca/wp-content/uploads/2021/01/To-Surveil-and-Predict1.1.pdf

    2020

  84. [92]

    Fei Pan, Sangryul Jeon, Brian Wang, Frank Mckenna, and Stella X. Yu. 2024. Zero-Shot Building Attribute Extraction From Large-Scale Vision and Language Models. 8647–8656. https://openaccess.thecvf.com/content/WACV2024/ html/Pan_Zero-Shot_Building_Attribute_Extraction_From_Larg...

    2024

  85. [93]

    Hauke Sandhaus, Angel Wang, Qian Yang, and Wendy Ju. 2024. My Precious Crash Data: Barriers and Opportunities in Encouraging Autonomous Driving Companies to Share Safety-Critical Data. arXiv:2504.17792 [cs.HC] https: //arxiv.org/abs/2504.17792 To appear in Proc. ACM Hum.-Compu...

    2024 arXiv

  86. [94]

    Shamier Settle and David Dyssegaard Kallick. 2024. Street Vendors of New York. Technical Report. Immigration Research Initiative. https://immresearch.org/ publications/street-vendors-of-new-york/

    2024

  87. [95]

    Shapira, Dorin, Franchi, Matthew, and Ju, Wendy. 2024. Fingerprinting New York City’s Scaffolding Problem with Longitudinal Dashcam Data

    2024

  88. [96]

    Yan Shvartzshnaider, Noah Apthorpe, Nick Feamster, and Helen Nissenbaum

  89. [97]

    Paul D Rosero-Montalvo, Diego Hernn Peluffo-Ordonez, Vivian Felix Lopez Batista, Jorge Serrano, and Edwin A Rosero. 2018. Intelligent system for identi- fication of wheelchair user’s posture using machine learning techniques. IEEE Sensors Journal 19, 5 (2018), 1936–1942

    2018

  90. [98]

    Luke Stark and Jesse Hoey. 2021. The ethics of emotion in artificial intelligence systems. In Proceedings of the 2021 ACM conference on fairness, accountability, and transparency. 782–793

    2021

  91. [99]

    Dodai Stewart and Juan Arredondo. 2024. Can New York City Street Vendors Survive a Police Crackdown? The New York Times (Sept. 2024). https://www. nytimes.com/2024/09/16/nyregion/street-wars-vendors.html

    2024

  92. [100]

    2017.Group privacy: New challenges of data technologies (1 ed.)

    Linnet Taylor, Luciano Floridi, and Bart van der Sloot (Eds.). 2017.Group privacy: New challenges of data technologies (1 ed.). Springer International Publishing, Cham, Switzerland. https://doi.org/10.1007/978-3-319-46608-8

    2017 doi

  93. [101]

    William Thackway, Matthew Ng, Chyi-Lin Lee, and Christopher Pettit. 2023. Implementing a deep-learning model using Google street view to combine social and physical indicators of gentrification. Computers, Environment and Urban Systems 102 (2023), 101970. https://www.sciencedi...

    2023

  94. [102]

    Shengbang Tong, Ellis Brown, Penghao Wu, Sanghyun Woo, Manoj Middepogu, Sai Charitha Akula, Jihan Yang, Shusheng Yang, Adithya Iyer, Xichen Pan, Austin Wang, Rob Fergus, Yann LeCun, and Saining Xie. 2024. Cambrian-1: A Fully Open, Vision-Centric Exploration of Multimodal LLMs....

    2024 arXiv

  95. [103]

    Corey Snyder and Minh Do. 2019. STREETS: A Novel Camera Network Dataset for Traffic Flow. In Advances in Neural Information Processing Sys- tems, Vol. 32. Curran Associates, Inc. https://proceedings.neurips.cc/paper/ 2019/hash/ee389847678a3a9d1ce9e4ca69200d06-Abstract.html

    2019

  96. [104]

    Gavrila, and Peter H

    Ries Uittenbogaard, Clint Sebastian, Julien Vijverberg, Bas Boom, Dariu M. Gavrila, and Peter H. N. de With. 2019. Privacy Protection in Street- View Panoramas Using Depth and Multi-View Imagery. 10581–10590. https://openaccess.thecvf.com/content_CVPR_2019/html/Uittenbogaard_ ...

    2019

  97. [105]

    Chiara Ullstein, Severin Engelmann, Orestis Papakyriakopoulos, Michel Ho- hendanner, and Jens Grossklags. 2022. AI-competent individuals and laypeople tend to oppose facial analysis AI. In Proceedings of the 2nd ACM Conference on Equity and Access in Algorithms, Mechanisms, an...

    2022

  98. [106]

    Chiara Ullstein, Severin Engelmann, Orestis Papakyriakopoulos, Yuko Ikkatai, Naira Paola Arnez-Jordan, Rose Caleno, Brian Mboya, Shuichiro Higuma, Tilman Hartwig, Hiromi Yokoyama, et al. 2024. Attitudes Toward Facial Analysis AI: A Cross-National Study Comparing Argentina, Ken...

    2024

  99. [107]

    Bart van der Sloot. 2017. Do groups have a right to protect their group interest in privacy and should they? Peeling the onion of rights and interests protected under article 8 ECHR. In Group Privacy. Springer International Publishing, Cham, 197–224. https://doi.org/10.1007/97...

    2017 doi

  100. [108]

    Anton Vedder. 1999. KDD: The challenge to individualism. Ethics and in- formation technology 1, 4 (Dec. 1999), 275–281. https://doi.org/10.1023/a: 1010016102284

    1999 doi

  101. [109]

    Dai Quoc Tran, Minsoo Park, Yuntae Jeon, Jinyeong Bak, and Seunghee Park

  102. [110]

    Li Yin, Qimin Cheng, Zhenxin Wang, and Zhenfeng Shao. 2015. ‘Big data’ for pedestrian volume: Exploring the use of Google Street View images for pedestrian counts. Applied Geography 63 (Sept. 2015), 337–345. https://doi.org/ 10.1016/j.apgeog.2015.07.010

    2015 doi

  103. [111]

    deployment

    J.D. Zamfirescu-Pereira, Jerry Chen, Emily Wen, Allison Koenecke, Nikhil Garg, and Emma Pierson. 2022. Trucks Don’t Mean Trump: Diagnosing Human Er- ror in Image Analysis. In Proceedings of the 2022 ACM Conference on Fairness, Accountability, and Transparency (FAccT ’22). Asso...

    2022

  104. [116]

    Jiahe Wang, Jiale Huang, Bingzhao Cai, Yifan Cao, Xin Yun, and Shangfei Wang

  105. [117]

    https://doi.org/10.48550/arXiv.2403.11450 arXiv:2403.11450

    Zero-shot Compound Expression Recognition with Visual Language Model at the 6th ABAW Challenge. https://doi.org/10.48550/arXiv.2403.11450 arXiv:2403.11450

    doi arXiv

  106. [467]

    https://www.cv-foundation.org/openaccess/content_cvpr_workshops_ 2014/W14/html/Leach_Detecting_Social_Groups_2014_CVPR_paper.html

    2014

  107. [2016]

    BMC Public Health 16, 1 (May 2016), 442

    Alcohol in urban streetscapes: a comparison of the use of Google Street View and on-street observation. BMC Public Health 16, 1 (May 2016), 442. https://doi.org/10.1186/s12889-016-3115-9

    2016 doi

  108. [2017]

    The ANNALS of the American Academy of Political and Social Science 669, 1 (Jan

    The Promise, Practicalities, and Perils of Virtually Auditing Neighbor- hoods Using Google Street View. The ANNALS of the American Academy of Political and Social Science 669, 1 (Jan. 2017), 18–40. https://doi.org/10.1177/ 0002716216681488 Publisher: SAGE Publications Inc

    2017

  109. [2018]

    https: //doi.org/10.48550/arXiv.1809.02236 arXiv:1809.02236 [cs]

    Analyzing Privacy Policies Using Contextual Integrity Annotations. https: //doi.org/10.48550/arXiv.1809.02236 arXiv:1809.02236 [cs]

    doi arXiv

  110. [2022]

    IEEE Access 10 (2022), 66061–66071

    Forest-Fire Response System Using Deep-Learning-Based Approaches With CCTV Images and Weather Data. IEEE Access 10 (2022), 66061–66071. https://doi.org/10.1109/ACCESS.2022.3184707 Conference Name: IEEE Access

    2022

  111. [2024]

    Proceedings of the ACM on Human-Computer Interaction 8, CSCW2 (2024), 1–23

    Privacy for Groups Online: Context Matters. Proceedings of the ACM on Human-Computer Interaction 8, CSCW2 (2024), 1–23

    2024

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.