Pith. sign in

REVIEW 3 major objections 5 minor 53 references

MITO: A Millimeter-Wave Dataset and Simulator for Non-Line-of-Sight Perception

T0 review · 3 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read The paper establishes MITO, a millimeter-wave dataset and simulator that enable non-line-of-sight perception of everyday objects.

desk verdict A genuinely useful public mmWave dataset and simulator for everyday-object NLOS perception, with an honest sim2real result; the headline simulator F-score is partly a fit and trajectory accuracy is unreported. read the letter →

arxiv 2502.10259 v3 pith:ZVVWSDL2 submitted 2025-02-14 cs.CV

Laura Dodds , Tara Boroushaki , Cusuh Ham , Fadel Adib This is my paper · ORCID
classification cs.CV
keywords millimeter-waveradarnon-line-of-sightperceptionsyntheticaperturemmWavedatasetsim-to-realtransferobjectsegmentationclassificationroboticmanipulation
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 presents MITO, a dataset of millimeter-wave (mmWave) radar images of 76 everyday objects, captured both in direct view and hidden inside cardboard or fabric, together with an open-source simulator that turns any 3D triangle mesh into matching synthetic mmWave images. The central claim is that these two pieces together give computer vision a workable path to non-line-of-sight perception: seeing and recognizing objects that visible-light cameras cannot. The authors demonstrate that the simulator matches real captures with a median 3D F-score of 94% (Table 1), and that a classifier trained only on synthetic images reaches 85.2% overall accuracy on real images, including 81.2% on fully occluded objects (Table 3). If correct, this means robots could verify packaged orders, retrieve occluded items, and classify hidden objects using signals that penetrate common packaging materials.

What carries the argument

The load-bearing identity is the coherent SAR sum $I(x,y,z)=\sum_{k=1}^{K}\sum_{j=1}^{N}S_{j,k}\,e^{j2\pi d(x,y,z,k)/\lambda_j}$ (Eq. 1), which adds up low-resolution radar frames from many antenna positions into one high-resolution complex image, with the phase term $2\pi d/\lambda$ depending on the round-trip distance from each antenna to each voxel. The simulator feeds this same pipeline with predicted reflections from a mesh: for each radar location it finds visible vertices, assigns each a phase $s_j(l,v)=e^{-j4\pi |l-v|/\lambda_j}$, and optionally suppresses reflections whose surface normal points away from the radar (specular model, Eq. 6) or keeps only edge vertices (edge model, Eq. 7); the two synthetic images are then blended as $I_{\mathrm{syn}}(\alpha_1,\alpha_2)=\frac{\alpha_1}{\alpha_1+\alpha_2}I_s+\frac{\alpha_2}{\alpha_1+\alpha_2}I_e$ (Eq. 8) to cover different materials.

What would settle it

Compare the UR5e's recorded poses against an independent motion-capture system while the arm moves the radar; if the position error exceeds roughly a millimeter, the 77 GHz synthetic-aperture images would visibly defocus, and both the simulator's 94% median F-score and the classifier's 85.2% accuracy would drop. Re-running the classification benchmark with a different trajectory or aperture size would also reveal how sensitive the claim is to trajectory accuracy.

Watch

Extended reading notes

Core claim

The paper's central discovery is that coherently summing the complex-valued reflections from a radar moving along a known robot trajectory - synthetic aperture radar, Eq. (1) - yields high-resolution images of everyday objects even when they sit inside a closed cardboard box, and that a lightweight simulator which predicts only two mechanisms of reflection, specular surface returns (Eq. 6) and edge diffraction (Eq. 7), reproduces these images closely enough to train a real-world classifier. With both simulation types blended by random weights during training, the classifier achieves 88.8% accuracy in line-of-sight and 81.2% in non-line-of-sight, for an overall 85.2% on real images (Table 3), and the simulator's best weighted output reaches a median 3D F-score of 94% against real point clouds (Table 1).

Load-bearing premise

Coherent focusing assumes the radar's position at each measurement is known to within a small fraction of the 3.9-millimeter wavelength at 77 GHz, and the paper relies on interpolating timestamped robot poses after network-time synchronization without reporting the resulting trajectory or synchronization error.

Editorial extensions

If this is right

  • Robotic systems can segment and classify objects inside closed cardboard boxes using mmWave images alone, with non-line-of-sight accuracy close to their line-of-sight accuracy.
  • Any 3D triangle mesh - from object dataset or phone scan - becomes training data for mmWave perception, removing the need for expensive real-world radar capture for each new object.
  • Fusing 24 GHz and 77 GHz images improves segmentation over either band alone, indicating that frequency diversity is a first-class design lever for through-occlusion perception.
  • The two-mechanism reflection model (specular plus edge) with random weight blending acts as a material augmentation, letting a network trained on synthetic images generalize to real objects of unmeasured material properties.
  • The dataset supplies a benchmark for two new vision tasks - NLOS object segmentation and NLOS shape classification - that are impossible for RGB-D cameras alone.

Reading between the lines

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

  • The paper's simulator models geometry only; objects whose reflections are dominated by volume scattering or strong absorption (for example, very wet or rough materials) may fall outside the two-mechanism model, so the claimed sim-to-real match is best understood as demonstrated on the 76 tested objects.
  • If the simulator generalizes as claimed, synthetic-aperture mmWave imaging becomes a standard pipeline for through-occlusion perception: generate any object's mesh, render synthetic radar images, and train a classifier without physical data collection.
  • A natural testable extension is classification accuracy as a function of occluder thickness and material; the paper tests cardboard and four layers of fabric, but the simulator does not model attenuation through occluders, so very thick or highly reflective packaging should degrade 77 GHz performance first.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The paper presents MITO, a dataset of millimeter-wave (mmWave) synthetic-aperture-radar (SAR) images of everyday objects, collected with a robotic arm carrying two radars (24 GHz and 77 GHz) and an RGB-D camera. The dataset contains over 24 million raw frames processed into 550 high-resolution, complex-valued 3D images of 76 YCB objects in both line-of-sight (LOS) and non-line-of-sight (NLOS) settings, together with RGB-D images, segmentation masks, and raw signals. The paper also introduces an open-source simulator that renders synthetic mmWave images from arbitrary 3D meshes using specular and edge-reflection models, and reports a median 3D F-score of 94% when comparing simulated and real point clouds. Two downstream applications are demonstrated: NLOS object segmentation using SAM with power-based prompting, and shape classification trained exclusively on synthetic images, achieving 85.2% overall accuracy on real LOS and NLOS images. The paper argues that MITO is the first diverse everyday-object mmWave dataset and that the simulator produces data matching real-world measurements, enabling broader NLOS perception.

Significance. If the dataset and simulator are validated, this work would be a significant contribution: it is the first open mmWave dataset of diverse everyday objects at multiple frequencies with raw data and code, and the sim2real classification result is a strong, independent demonstration that the simulator captures enough physical fidelity to be useful for training perception models. The segmentation benchmark provides a concrete baseline for NLOS perception, and the open-source simulator could enable large-scale synthetic data generation for future research. However, the headline simulator accuracy is weakened by the fact that it is computed after per-object fitting of the combination weights (Eq. 11), and the entire dataset rests on an unvalidated assumption about antenna-position accuracy for coherent SAR focusing. These issues must be resolved before the claims of 'matches real-world collected data' can be accepted as stated.

major comments (3)
  1. [Sec. 5.2.2, Eq. (11)] The reported simulator F-scores, including the median of 94% in Table 1, are computed after selecting the best (α1, α2) weights for each object individually via Eq. (11). This makes the result a measure of the simulator's representational flexibility, not its predictive accuracy: the weights are fit to the same real images used for evaluation. The claim in Sec. 2 that the simulator 'generates synthetic data that matches real-world collected data' is therefore overstated. Please report F-scores for fixed weights (e.g., equal weights), or use a cross-validation scheme where weights are selected on a training subset and evaluated on held-out objects, and include the distribution of the selected weights.
  2. [Sec. 4.3, Eq. (1)] Coherent SAR focusing via Eq. (1) requires accurate antenna positions p_k in the phase term exp(j2π d/λ). At 77 GHz (λ ≈ 3.9 mm), a position error of just a few tenths of a millimeter produces a phase error on the order of a radian, defocusing the image. The paper states that robot locations are interpolated to radar timestamps after NTP synchronization between two computers, but it reports no trajectory accuracy, synchronization offset/jitter, or interpolation error. Without a quantitative validation of the antenna-position error (e.g., a calibration measurement with a point reflector, or a comparison of focused images against a ground-truth target), the physical validity of all 550 real SAR images—and hence every downstream result in Tables 2 and 3—is unestablished.
  3. [Sec. 5.2.1, Eqs. (9)-(10), Sec. 5.1.1-5.1.2] The simulator and its evaluation depend on several free thresholds: the specular angle threshold τ (Eq. 6), the edge angle threshold τ_e (Sec. 5.1.2), the point-cloud power threshold τ_P (Sec. 5.2.1), and the F-score distance threshold τ_F (Eqs. 9-10). No values are reported for these thresholds, and no sensitivity analysis is provided. Since the reported F-scores and the point-cloud comparisons can be expected to vary substantially with these choices, the simulator accuracy claim is under-specified. Please state the threshold values and show that the conclusions (e.g., the superior performance of the combined model) are robust to reasonable variations.
minor comments (5)
  1. [Abstract] The phrase 'non-light-of-sight' should be 'non-line-of-sight' for consistency with the rest of the text.
  2. [Sec. 5.1.3] The subsection title 'Combing Reflections' is a typo; it should be 'Combining Reflections'.
  3. [Table 2] The asterisk next to the camera NLOS recall value (100%*) is never explained in the caption or body text; please add a footnote explaining that the mask covers the occluder rather than the object.
  4. [Sec. 5.1.1] The threshold τ in Eq. (6) is described as being on the angle between the normal and the radar-to-vertex vector, but the condition is written without explicit units or a default value; please clarify the range and units of τ.
  5. [Sec. 4.2.1] The paper mentions that the tripod/styrofoam background can be removed by subtracting an empty scene, but it does not state how many of the 550 images actually use this background subtraction. This is useful information for users of the dataset.

Circularity Check

1 steps flagged · score 4.0 of 10

Partial circularity: the simulator's headline 94% median F-score is obtained by choosing material weights that maximize F against the same real images (Eq. 11), while the sim2real classification provides independent, non-circular support.

  1. fitted input called prediction [Section 5.2.2, Eq. 11 and Table 1; Related Work Section 2]
    "Since our goal is not to select a single set of weights for each object, but to allow the simulation to represent many different objects, we compute this metric across a range of weights {α1, α2} and choose the combined simulation which produces the best F-score ... F = max_{(α1,α2)∈W} F(Isyn(α1, α2), I)"

    The simulator-accuracy headline (median 3D F-score of 94%, Table 1) is computed by optimizing the two material-combination weights per real object against that same real object's mmWave image (Eq. 11). The reported F-score is therefore a best-case upper envelope of a per-object fit, not the accuracy of a fixed, parameter-free simulator. The Related Work claim that the simulator 'generates synthetic data that matches real-world collected data' rests on this fitted value. The sim2real classifier (Sec. 6.2) does break the loop because it uses random weights and evaluates on held-out real images, but the Table 1 match claim itself is circular by construction: the metric being reported is maximized over the very data used to measure it.

full rationale

The only substantive circularity is the simulator validation in Sec. 5.2.2: Eq. 11 selects per-object weights α1, α2 that maximize the 3D F-score against the real image, and Table 1 then reports those maximized F-scores as evidence that the simulator matches reality. That is a fitted input being presented as a match accuracy, which is a partial circularity. The paper is otherwise self-contained: the SAR imaging pipeline (Eq. 1) is a direct physical model applied to independently collected radar frames; the dataset and NLOS segmentation benchmarks do not reduce to their inputs; and the sim2real classification (Sec. 6.2) provides genuinely independent evidence by training on synthetic images with randomized reflection weights and evaluating on real images never seen in training. The self-citations that appear (e.g., [18] for point-cloud generation, [10] for background subtraction) are methodological references, not load-bearing justifications of the central claims. Because the central dataset contribution and the sim2real evaluation have independent content, the overall circularity is partial rather than pervasive; a score of 4 reflects the one fitted simulator-accuracy claim while acknowledging the independent grounding elsewhere.

Assumptions & free parameters 6 free parameters · 6 assumptions · 0 invented entities

The central contributions rest on a small number of fitted thresholds and weights, plus domain assumptions about the physical reflection model, trajectory accuracy, and the validity of YCB meshes. No new physical entities are introduced.

free parameters (6)
  • alpha1, alpha2 combination weights = not reported; chosen per object
    In Eq. 11 the simulator accuracy F-score is the maximum over a weight grid for each real object, so the reported 94% median is a fitted value, not an independent prediction.
  • specular angle threshold tau = not reported
    Eq. 6 requires a threshold on surface-normal angle to decide whether a specular reflection returns to the radar; threshold value is not stated.
  • edge angle threshold tau_e = not reported
    Defines which vertices are edge vertices in Sec. 5.1.2; value not stated.
  • point cloud power threshold tau_P = not reported
    Used to convert mmWave images to point clouds before F-score; threshold value not stated.
  • F-score distance threshold tau_F = not reported
    Distance threshold in Eq. 9-10; value not stated.
  • SAM prompt point count and selection thresholds = not reported
    Segmentation selects a subset of highest-power reflection points as SAM prompts; count and power thresholds for multi-spectral fusion are not specified in the main text.
assumptions (6)
  • domain assumption Coherent SAR backprojection with NTP-synchronized antenna positions yields focused images.
    Eq. 1 requires position accuracy well below a wavelength (about 4 mm at 77 GHz); Sec. 4.3 does not validate trajectory error.
  • domain assumption Specular and edge-only scattering (Eqs. 4-7) captures dominant mmWave reflections for everyday objects.
    Sec. 5.1; diffuse reflections, multi-bounce, and material-dependent coefficients are not modeled.
  • domain assumption YCB meshes accurately represent the physical test objects.
    Used to generate simulator images and synthetic training data; physical objects may differ from meshes.
  • domain assumption SAM, trained on RGB images, transfers to colorized mmWave projections when prompted by high-power points.
    Sec. 6.1.1; no fine-tuning or adaptation is described.
  • domain assumption Human-in-the-loop SAM annotation of RGB-D images produces accurate ground-truth masks.
    Sec. 4.3; no verification of annotation accuracy is reported.
  • domain assumption Cardboard and cardboard-plus-fabric layers represent the full NLOS setting claimed in the abstract.
    Only these occluders are tested; plastic, wood, and other common occluders are not evaluated.

how reviews work

0 comments
Cite this review

Pith. "Pith review of MITO: A Millimeter-Wave Dataset and Simulator for Non-Line-of-Sight Perception." pith.science (2026). https://pith.science/paper/ZVVWSDL2

@misc{pith2026250210259,
  author       = {Pith},
  title        = {Pith review of: MITO: A Millimeter-Wave Dataset and Simulator for Non-Line-of-Sight Perception},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZVVWSDL2}},
  note         = {Machine review of arXiv:2502.10259}
}
read the original abstract

The ability to observe the world is fundamental to reasoning and making informed decisions on how to interact with the environment. However, optical perception can often be disrupted due to common occurrences, such as occlusions, which can pose challenges to existing vision systems. We present MITO, the first millimeter-wave (mmWave) dataset of diverse, everyday objects, collected using a UR5 robotic arm with two mmWave radars operating at different frequencies and an RGB-D camera. Unlike visible light, mmWave signals can penetrate common occlusions (e.g., cardboard boxes, fabric, plastic) but each mmWave frame has much lower resolution than typical cameras. To capture higher-resolution mmWave images, we leverage the robot's mobility and fuse frames over the synthesized aperture. MITO captures over 24 million mmWave frames and uses them to generate 550 high-resolution mmWave (synthetic aperture) images in line-of-sight and non-light-of-sight (NLOS), as well as RGB-D images, segmentation masks, and raw mmWave signals, taken from 76 different objects. We develop an open-source simulation tool that can be used to generate synthetic mmWave images for any 3D triangle mesh. Finally, we demonstrate the utility of our dataset and simulator for enabling broader NLOS perception by developing benchmarks for NLOS segmentation and classification.

Figures

Figures reproduced from arXiv: 2502.10259 by the authors.

Figure 1
Figure 1. MITO. We use a robotic arm to move mmWave radars through￾out the environment. While an RGB-D camera’s output cannot see inside the box, we produce high-resolution non-line-of-sight mmWave images. cannot see inside a closed box to confirm e-commerce or￾ders are correct or detect broken objects during shipping. Nor is it feasible for camera-based robots to efficiently plan complex, multi-step tasks in practical enviro… view at source ↗
Figure 2
Figure 2. mmWave Imaging a) mmWave radars estimate range and angle-of-arrival to produce b) re￾flection maps [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 4
Figure 4. Sample Images. A number of example mmWave images within MITO. DCA1000EVM [1]), 24 GHz radar (Infineon Posi￾tion2Go [4]), and Intel Realsense D415 [25] to an end￾effector of a UR5e robotic arm [44] using custom-designed 3D-printed parts. Since the robot and radars are controlled from different computers, we synchronize them by setting both computers to the same NTP time server. We place the objects in two setups that… view at source ↗
Figures from the paper (4 more)
Figure 5
Figure 5. Figure 5: Multi-Spectral Imaging. a-b) show 77 GHz & 24 GHz images for a padlock in LOS. c-d) show the images in NLOS. then use the camera’s location to transform and align the RGB-D ground-truth mask with the mmWave image. More details on the alignment process are in the supple…
Figure 6
Figure 6. Figure 6: Simulation Output. Our simulation uses a a) 3D mesh to produce images assuming different reflection types: b) specular & c) edge. d/e) We combine these with different weights to simulate different materials. Simulation 25th 50th 75th Combined 85% 94% 98% Specular 68% 9…
Figure 7
Figure 7. Figure 7: Segmentation a)Colorize 2D image b)Select prompt points c)Segment with SAM LOS NLOS Input Pre. Recall F1 IoU Acc. Pre. Recall F1 IoU Acc. Camera 99.7% 99.5% 99.4% 98.9% 99.9% 3.6% 100%* 7.0% 3.6% 14.0% 24 GHz 57.9% 50.1% 37.3% 22.9% 93.7% 53.0% 38.7% 29.1% 17.0% 92.1% …
Figure 8
Figure 8. Figure 8: Qualitative Segmentation Result. An example where the depth image is unable to capture the entire object, but the mmWave image does. Training Data Accuracy Edge Specular LOS NLOS Overall X 61.1% 75.0% 67.6% X 77.7% 75.0% 76.4% X X 88.8% 81.2% 85.2% [PITH_FULL_IMAGE:fi…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

53 extracted references · 50 canonical work pages

  1. [1]

    https : / / www

    Ti dca1000evm. https : / / www . ti . com / tool / DCA1000EVM. 4

    work page

  2. [2]

    https : / / www

    Icp registration. https : / / www . open3d . org / docs / release / tutorial / pipelines / icp _ registration.html. 6

    work page

  3. [3]

    https://ps.is

    Perceiving systems mesh package. https://ps.is. mpg.de/code/mesh-library. 5

    work page

  4. [4]

    https://www.infineon.com/ cms / en / product / evaluation - boards / demo - position2go/

    Infineon position2go. https://www.infineon.com/ cms / en / product / evaluation - boards / demo - position2go/. 4

    work page

  5. [5]

    https : / / www

    Understanding mmwave for 5g networks. https : / / www . 5gamericas . org / wp - content / uploads / 2020 / 12 / InDesign - Understanding - mmWave - for-5G-Networks.pdf. 3

    work page 2020

  6. [6]

    https://www.ti.com/product/ IWR1443#tech-docs

    Ti iwr1443boost. https://www.ti.com/product/ IWR1443#tech-docs. 1, 3

    work page

  7. [7]

    https://www

    mmWave Radar Radome Design Guide. https://www. ti . com / lit / an / swra705 / swra705 . pdf, 2021. Texas Instruments. 4

    work page 2021

  8. [8]

    An- sys HFSS

    Ansys high frequency structure simulator (hfss), 2022. An- sys HFSS. 2

    work page 2022

Show all 53 references
  1. [9]

    3d tracking via body radio reflections

    Fadel Adib, Zach Kabelac, Dina Katabi, and Robert C Miller. 3d tracking via body radio reflections. In 11th USENIX Sym- posium on Networked Systems Design and Implementation (NSDI 14), pages 317–329, 2014. 3

    2014

  2. [10]

    Capturing the human figure through a wall

    Fadel Adib, Chen-Yu Hsu, Hongzi Mao, Dina Katabi, and Fr´edo Durand. Capturing the human figure through a wall. ACM Transactions on Graphics (TOG), 34(6):219, 2015. 4

    2015

  3. [11]

    Seeing through fog without seeing fog: Deep multimodal sensor fu- sion in unseen adverse weather

    Mario Bijelic, Tobias Gruber, Fahim Mannan, Florian Kraus, Werner Ritter, Klaus Dietmayer, and Felix Heide. Seeing through fog without seeing fog: Deep multimodal sensor fu- sion in unseen adverse weather. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR...

    2020

  4. [12]

    Millinoise: a millimeter-wave radar sparse point cloud dataset in indoor scenarios

    Walter Brescia, Pedro Gomes, Laura Toni, Saverio Mascolo, and Luca De Cicco. Millinoise: a millimeter-wave radar sparse point cloud dataset in indoor scenarios. In Proceed- ings of the 15th ACM Multimedia Systems Conference, pages 422–428, 2024. 2

    2024

  5. [13]

    Caesar, V

    H. Caesar, V . Bankiti, A. H. Lang, S. V ora, V . Liong, Q. Xu, A. Krishnan, Y . Pan, G. Baldan, and O. Beijbom. nuscenes: A multimodal dataset for autonomous driving. In 2020 IEEE/CVF Conference on Computer Vision and Pat- tern Recognition (CVPR), pages 11618–11628, Los Alami...

    2020

  6. [14]

    Yale-cmu-berkeley dataset for robotic manipulation research

    Berk Calli, Arjun Singh, James Bruce, Aaron Walsman, Kurt Konolige, Siddhartha Srinivasa, Pieter Abbeel, and Aaron M Dollar. Yale-cmu-berkeley dataset for robotic manipulation research. The International Journal of Robotics Research , 36(3):261–268, 2017. 2, 3, 5

    2017

  7. [15]

    mmbody benchmark: 3d body re- construction dataset and analysis for millimeter wave radar,

    Anjun Chen, Xiangyu Wang, Shaohao Zhu, Yanxu Li, Jim- ing Chen, and Qi Ye. mmbody benchmark: 3d body re- construction dataset and analysis for millimeter wave radar,

  8. [16]

    Milipoint: A point cloud dataset for mmwave radar, 2023

    Han Cui, Shu Zhong, Jiacheng Wu, Zichao Shen, Naim Dah- noun, and Yiren Zhao. Milipoint: A point cloud dataset for mmwave radar, 2023. 2

    2023

  9. [17]

    Tera- hertz signal transparent relay and switching using photonic technology

    Pham Tien Dat, Yuya Yamaguchi, Shingo Takano, Shotaro Hirata, Junichiro Ichikawa, Ryo Shimizu, Keizo Inagaki, Isao Morohashi, Yuki Yoshida, Atsushi Kanno, Naokatsu Ya- mamoto, Tetsuya Kawanishi, and Kouichi Akahane. Tera- hertz signal transparent relay and switching using phot...

    2024

  10. [18]

    Around the corner mmwave imaging in practical environments

    Laura Dodds, Hailan Shanbhag, Junfeng Guan, Saurabh Gupta, and Haitham Hassanieh. Around the corner mmwave imaging in practical environments. In Proceedings of the 30th Annual International Conference on Mobile Computing and Networking, pages 1–15, 2024. 6

    2024

  11. [19]

    Everingham, L

    M. Everingham, L. Van Gool, C. K. I. Williams, J. Winn, and A. Zisserman. The pascal visual object classes (voc) challenge. International Journal of Computer Vision, 88(2): 303–338, 2010. 2

    2010

  12. [20]

    Frequency Allo- cation:

    Federal Communications Commission. Frequency Allo- cation:. transition.fcc.gov/oet/spectrum/ table/fcctable.pdf. 3

  13. [21]

    Instructing people for training gestural in- teractive systems

    Simon Fothergill, Helena Mentis, Pushmeet Kohli, and Se- bastian Nowozin. Instructing people for training gestural in- teractive systems. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, page 1737–1746, New York, NY , USA, 2012. Association for Co...

    2012

  14. [22]

    Through fog high-resolution imag- ing using millimeter wave radar

    Junfeng Guan, Sohrab Madani, Suraj Jog, Saurabh Gupta, and Haitham Hassanieh. Through fog high-resolution imag- ing using millimeter wave radar. In IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020. 2

    2020

  15. [23]

    Talk2radar: Bridging natural language with 4d mmwave radar for 3d referring expression comprehension

    Runwei Guan, Ruixiao Zhang, Ningwei Ouyang, Jianan Liu, Ka Lok Man, Xiaohao Cai, Ming Xu, Jeremy Smith, Eng Gee Lim, Yutao Yue, et al. Talk2radar: Bridging natural language with 4d mmwave radar for 3d referring expression comprehension. arXiv preprint arXiv:2405.12821, 2024. 2

    2024 arXiv

  16. [24]

    Yalavarthy, and Linga Reddy Cenkeramaddi

    Siddharth Gupta, Prabhat Kumar Rai, Abhinav Kumar, Pha- neendra K. Yalavarthy, and Linga Reddy Cenkeramaddi. Tar- get classification by mmwave fmcw radars using machine learning on range-angle images. IEEE Sensors Journal, 21 (18):19993–20001, 2021. 2

    2021

  17. [25]

    https://www.intelrealsense

    Intel RealSense. https://www.intelrealsense. com, 2019. 4

    2019

  18. [26]

    Barron, Mario Fritz, Kate Saenko, and Trevor Darrell

    Allison Janoch, Sergey Karayev, Yangqing Jia, Jonathan T. Barron, Mario Fritz, Kate Saenko, and Trevor Darrell. A category-level 3-d object dataset: Putting the kinect to work. In 2011 IEEE International Conference on Computer Vision Workshops (ICCV Workshops), pages 1168–1174...

    2011

  19. [27]

    Berg, Wan-Yen Lo, Piotr Doll ´ar, and Ross Girshick

    Alexander Kirillov, Eric Mintun, Nikhila Ravi, Hanzi Mao, Chloe Rolland, Laura Gustafson, Tete Xiao, Spencer White- head, Alexander C. Berg, Wan-Yen Lo, Piotr Doll ´ar, and Ross Girshick. Segment anything. arXiv:2304.02643, 2023. 2, 4, 6, 7

    2023 arXiv

  20. [28]

    Coloradar: The direct 3d millimeter wave radar dataset

    Andrew Kramer, Kyle Harlow, Christopher Williams, and Christoffer Heckman. Coloradar: The direct 3d millimeter wave radar dataset. The International Journal of Robotics Research, 41(4):351–360, 2022. 2

    2022

  21. [29]

    Classification of targets detected by mmwave radar using yolov5

    Mohamed Lamane, Mohamed Tabaa, and Abdessamad Klilou. Classification of targets detected by mmwave radar using yolov5. Procedia Computer Science , 203:426–431,

  22. [30]

    Concealed object detection and recognition sys- tem based on millimeter wave fmcw radar.Applied Sciences, 11(19), 2021

    Jie Liu, Kai Zhang, Zhenlin Sun, Qiang Wu, Wei He, and Hao Wang. Concealed object detection and recognition sys- tem based on millimeter wave fmcw radar.Applied Sciences, 11(19), 2021. 2

    2021

  23. [31]

    Concealed object detection for activate millimeter wave image

    Ting Liu, Yao Zhao, Yunchao Wei, Yufeng Zhao, and Shikui Wei. Concealed object detection for activate millimeter wave image. IEEE Transactions on Industrial Electronics, 66(12): 9909–9917, 2019. 2

    2019

  24. [32]

    Lu, Patrick Cabrol, Daniel Steinbach, and Ravikumar V

    Jonathan S. Lu, Patrick Cabrol, Daniel Steinbach, and Ravikumar V . Pragada. Measurement and characterization of various outdoor 60 ghz diffracted and scattered paths. In MILCOM 2013 - 2013 IEEE Military Communications Con- ference, pages 1238–1243, 2013. 3

    2013

  25. [33]

    End-to-end simulation of 5g mmwave networks

    Marco Mezzavilla, Menglei Zhang, Michele Polese, Russell Ford, Sourjya Dutta, Sundeep Rangan, and Michele Zorzi. End-to-end simulation of 5g mmwave networks. IEEE Com- munications Surveys & Tutorials, 20(3):2237–2263, 2018. 2

    2018

  26. [34]

    K-radar: 4d radar object detection for autonomous driving in various weather conditions

    Dong-Hee Paek, Seung-Hyun Kong, and Kevin Tirta Wijaya. K-radar: 4d radar object detection for autonomous driving in various weather conditions. In Thirty-sixth Conference on Neural Information Processing Systems Datasets and Bench- marks Track, 2022. 2

    2022

  27. [35]

    Andras Palffy, Ewoud Pool, Srimannarayana Baratam, Ju- lian F. P. Kooij, and Dariu M. Gavrila. Multi-class road user detection with 3+1d radar in the view-of-delft dataset. IEEE Robotics and Automation Letters, 7(2):4961–4968, 2022. 2

    2022

  28. [36]

    Radarhd: Demonstrating lidar-like point clouds from mmwave radar

    Akarsh Prabhakara, Tao Jin, Arnav Das, Gantavya Bhatt, Lilly Kumari, Elahe Soltanaghai, Jeff Bilmes, Swarun Ku- mar, and Anthony Rowe. Radarhd: Demonstrating lidar-like point clouds from mmwave radar. InProceedings of the 29th Annual International Conference on Mobile Computin...

    2023

  29. [37]

    Mmvr: Millimeter-wave multi-view radar dataset and benchmark for indoor perception

    M Mahbubur Rahman, Ryoma Yataka, Sorachi Kato, Pu Perry Wang, Peizhao Li, Adriano Cardace, and Pet- ros Boufounos. Mmvr: Millimeter-wave multi-view radar dataset and benchmark for indoor perception. arXiv preprint arXiv:2406.10708, 2024. 2

    2024 arXiv

  30. [38]

    Squigglemilli: Approximating sar imaging on mobile millimeter-wave devices

    Hem Regmi, Moh Sabbir Saadat, Sanjib Sur, and Srihari Nelakuditi. Squigglemilli: Approximating sar imaging on mobile millimeter-wave devices. Proc. ACM Interact. Mob. Wearable Ubiquitous Technol., 5(3), 2021. 2

    2021

  31. [39]

    Object recognition for mil- limeter wave mimo-sar images based on high-resolution fea- ture recursive alignment fusion network.IEEE Sensors Jour- nal, 23(14):16413–16427, 2023

    Bofeng Su and Minghui Yuan. Object recognition for mil- limeter wave mimo-sar images based on high-resolution fea- ture recursive alignment fusion network.IEEE Sensors Jour- nal, 23(14):16413–16427, 2023. 2

    2023

  32. [40]

    Multi-source aggregation transformer for concealed object detection in millimeter-wave images

    Peng Sun, Ting Liu, Xiaotong Chen, Shiyin Zhang, Yao Zhao, and Shikui Wei. Multi-source aggregation transformer for concealed object detection in millimeter-wave images. IEEE Transactions on Circuits and Systems for Video Tech- nology, 32(9):6148–6159, 2022. 2

    2022

  33. [41]

    A novel millimeter-wave channel simulator and appli- cations for 5g wireless communications

    Shu Sun, George R MacCartney, and Theodore S Rappa- port. A novel millimeter-wave channel simulator and appli- cations for 5g wireless communications. In2017 IEEE inter- national conference on communications (ICC) , pages 1–7. IEEE, 2017. 2

    2017

  34. [42]

    Y . Sun, Z. Huang, H. Zhang, Z. Cao, and D. Xu. 3drimr: 3d reconstruction and imaging via mmwave radar based on deep learning. In 2021 IEEE International Performance, Comput- ing, and Communications Conference (IPCCC) , pages 1–8, Los Alamitos, CA, USA, 2021. IEEE Computer Society. 2

    2021

  35. [43]

    Richter*, Ren ´e Ranftl, Zhuwen Li, Vladlen Koltun, and Thomas Brox

    Maxim Tatarchenko*, Stephan R. Richter*, Ren ´e Ranftl, Zhuwen Li, Vladlen Koltun, and Thomas Brox. What do single-view 3d reconstruction networks learn? 2019. 6

    2019

  36. [44]

    https://www.universal- robots.com/products/ur5-robot/, 2021

    Universal Robots, UR5e. https://www.universal- robots.com/products/ur5-robot/, 2021. 4

    2021

  37. [45]

    Vision meets mmwave radar: 3d object perception benchmark for autonomous driving

    Yizhou Wang, Jen-Hao Cheng, Jui-Te Huang, Sheng-Yao Kuan, Qiqian Fu, Chiming Ni, Shengyu Hao, Gaoang Wang, Guanbin Xing, Hui Liu, et al. Vision meets mmwave radar: 3d object perception benchmark for autonomous driving. In 2024 IEEE Intelligent Vehicles Symposium (IV), pages 2769–

    2024

  38. [46]

    Wrdi: A multimodal dataset of mmwave radar data and im- age

    Zhongce Wang, Chao Jin, Lingzhang Sun, and Yunfan Li. Wrdi: A multimodal dataset of mmwave radar data and im- age. In 2021 IEEE International Conference on Big Data (Big Data), pages 3209–3214, 2021. 2

    2021

  39. [47]

    A simulation-based framework for the design of human activity recognition sys- tems using radar sensors

    Sahil Waqar and Matthias P ¨atzold. A simulation-based framework for the design of human activity recognition sys- tems using radar sensors. IEEE Internet of Things Journal ,

  40. [48]

    mmhpe: Robust multi-scale 3d human pose estimation using a single mmwave radar

    Yingxiao Wu, Zhongmin Jiang, Haocheng Ni, Changlin Mao, Zhiyuan Zhou, Wenxiang Wang, and Jianping Han. mmhpe: Robust multi-scale 3d human pose estimation using a single mmwave radar. IEEE Internet of Things Journal ,

  41. [49]

    Autonomous mining transportation systems: Integrating 4d mmwave radar for en- hanced detection of obstructed static objects

    Jianjian Yang, Tianmu Gui, Yibo Tong, Yuyuan Zhang, Qiankun Huang, and Guanghui Zhao. Autonomous mining transportation systems: Integrating 4d mmwave radar for en- hanced detection of obstructed static objects. IEEE Transac- tions on Intelligent Vehicles, 2024. 2

    2024

  42. [50]

    Synthesized millimeter-waves for human motion sensing

    Xiaotong Zhang, Zhenjiang Li, and Jin Zhang. Synthesized millimeter-waves for human motion sensing. In Proceed- ings of the 20th ACM Conference on Embedded Networked Sensor Systems, page 377–390, New York, NY , USA, 2023. Association for Computing Machinery. 2

    2023

  43. [51]

    Tj4dradset: A 4d radar dataset for autonomous driving

    Lianqing Zheng, Zhixiong Ma, Xichan Zhu, Bin Tan, Sen Li, Kai Long, Weiqi Sun, Sihan Chen, Lu Zhang, Mengyue Wan, Libo Huang, and Jie Bai. Tj4dradset: A 4d radar dataset for autonomous driving. In 2022 IEEE 25th International Con- ference on Intelligent Transportation Systems ...

    2022

  44. [52]

    4d radar simulator for human activity recognition

    Junyu Zhou and Julien Le Kernec. 4d radar simulator for human activity recognition. IET Radar, Sonar & Navigation,

  45. [2022]

    17th International Conference on Future Networks and Communications / 19th International Conference on Mo- bile Systems and Pervasive Computing / 12th International Conference on Sustainable Energy Information Technology (FNC/MobiSPC/SEIT 2022), August 9-11, 2022, Niagara Fall...

    2022

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.