Pith. sign in

REVIEW 4 major objections 6 minor 70 references

Recognizing Part Attributes with Insufficient Data

T0 review · 4 major / 6 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read The paper argues that a part attribute decomposes into part location plus appearance pattern, and that sharing these concepts lets a network recognize attributes with few or zero training samples.

desk verdict A genuinely useful factored-sharing idea for scarce part-attribute labels, with a solid few-shot story but a zero-shot claim that is stronger than the evidence. read the letter →

arxiv 1908.03335 v2 pith:GD2CJ3IY submitted 2019-08-09 cs.CV

Xiangyun Zhao , Yi Yang , Feng Zhou , Xiao Tan , Yuchen Yuan , Yingze Bao , Ying Wu This is my paper · ORCID
classification cs.CV
keywords partattributerecognitionconceptsharingzero-shotlearningvisualattentiondatainsufficiencyfine-grainedhumandatasetfew-shot
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

This paper claims that a part attribute—say “wing white”—is best treated not as one indivisible label but as the combination of two learnable concepts: where the part sits in the image and what the appearance pattern looks like. It introduces the Concept Sharing Network (CSN), which learns a shared localization module per part and a shared pattern classifier per appearance, so every attribute label contributes gradients to two reusable modules instead of one isolated branch. Because each module trains on the union of all attributes that share its part or pattern, the effective training data for each concept is much larger than any single attribute's label count. The payoff is that attributes with very few positive samples improve, and an attribute never present in training—a new part-pattern pair such as “wing red”—can still be recognized by combining already learned modules. Experiments on bird, face, and surveillance-human data show gains that grow as per-attribute data shrinks.

What carries the argument

The named object is the Concept Sharing Network (CSN), built from part-localization modules and appearance-pattern modules that are shared across attribute labels. For part $i$, a learnable representation $V_i$ is multiplied against the CNN feature map $Q(x;\Theta)$, normalized by spatial softmax, and used to weight the features so the pooled representation focuses on the part; for pattern $j$, a shared binary classifier $W_j$ is applied to that pooled, attended feature vector. The load-bearing detail is the update rule: the gradient for $V_i$ sums over all samples whose attribute uses part $i$, and the gradient for $W_j$ sums over all samples whose attribute uses pattern $j$, which converts many small per-attribute datasets into two larger per-concept datasets.

What would settle it

Build a benchmark where the same pattern word covers visually unlike textures on different parts—e.g. “striped” on a bird's bill versus its wing—train CSN only on the source part, and measure average precision on the unseen part; if it is no better than the no-sharing baseline, the cross-part pattern transfer that zero-shot recognition depends on is falsified.

Watch

Extended reading notes

Core claim

The central claim is that the recognition model for attribute $P_{i,j}$ is fully determined by three parameter sets: the shared CNN features $\Theta$, a part-localization vector $V_i$, and a pattern-classifier weight $W_j$. During training, $V_i$ is updated by gradients from every attribute whose part is $i$, and $W_j$ from every attribute whose pattern is $j$, while the loss for each sample uses whichever pair matches its label. Therefore the model for $P_{i,j}$ is determined even when $P_{i,j}$ itself has zero training data, because $W_j$ can be learned from $P_{\alpha,j}$ ($\alpha\neq i$) and $V_i$ can be learned from $P_{i,\beta}$ ($\beta\neq j$). The paper shows empirically that this sharing improves average precision most for attributes with few samples, and that zero-shot combinations reach accuracy comparable to supervised training for several attributes.

Load-bearing premise

The load-bearing premise is that an appearance pattern such as “grey” or “spotted” looks similar enough across different object parts that one shared pattern classifier trained on some parts transfers to a new part—a premise the paper itself notes is imperfect, since pattern features vary across locations.

Editorial extensions

If this is right

  • Attributes with very few positive training samples benefit more from concept sharing than attributes with abundant samples, so the method is aimed at long-tailed label distributions.
  • Part-location annotation is not needed; image-level attribute labels alone can localize parts through attention and classify their appearance.
  • A part attribute never present in training can be recognized by combining a learned part-location module with a learned appearance-pattern module.
  • Increasing the number of attributes sharing a part module steadily improves accuracy for the tested attributes.
  • On the surveillance-human dataset, sharing raises mean average precision from 30.3% to 51.2%.

Reading between the lines

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

  • If the transferability condition holds, the same part-plus-pattern decomposition should apply to other compositional visual labels, such as material-and-color pairs or object-action combinations, wherever the same appearance concept recurs on multiple locations.
  • The zero-shot comparison in the paper mixes attributes with very different supporting-data sizes; a fairer audit would stratify zero-shot attributes by how much data their pattern and part modules had, predicting that the largest zero-shot gains occur when both modules are well supplied.
  • The mechanism suggests a direct data-efficiency law: the gain from sharing should increase with the ratio of shared-module training data to per-attribute training data, a prediction testable by plotting AP gain against that ratio across attributes.
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 proposes a Concept Sharing Network (CSN) for part attribute recognition, in which each part attribute is modeled as the combination of a part-localization module (a learnable attention vector per part) and an appearance-pattern module (a binary classifier per pattern). These modules are shared across attributes, so that the data for one attribute contributes to learning the part and pattern modules used by other attributes. The authors claim this improves recognition especially when per-attribute training data are scarce, and enables zero-shot recognition of attribute combinations never seen in training. Experiments are reported on CUB-200-2011, CelebA, and a newly introduced human attribute dataset (SurveilA).

Significance. The central idea is clean and well motivated: decomposing a part attribute into part location and appearance pattern, then sharing those concepts across attributes, is a plausible mechanism for data-efficient attribute learning. The formal setup in Sec. 3.1-3.3 is coherent, and the sharing updates in Eqs. 6-7 correctly enlarge the effective training set for each module. The paper also contributes a large new surveillance benchmark (SurveilA) and promises to release code, which are useful to the community. If the zero-shot claim holds generally, the work would be a meaningful step toward attribute recognition under extreme data scarcity. However, the current evaluation does not yet establish that generality: the evidence is drawn from selected attribute subsets, a single zero-shot split, and no error bars, and one reported zero-shot row directly contradicts the transfer premise.

major comments (4)
  1. [Sec. 4.3.5, Table 8] The zero-shot evaluation rests on a single random split of 20 unseen attributes with no repeated runs or confidence intervals, and the paper does not report which 20 attributes were held out or whether the split is available for comparison. More importantly, the row for 'bill yellow' shows zero-shot AP of 9.2% versus 50.3% for supervised training, despite 215 positive samples for that attribute and abundant 'bill' and 'yellow' concepts elsewhere in the training set. This is a direct counterexample to the claim that CSN can recognize unseen attributes by recombining learned part and pattern modules, and the paper reports it without analysis. The authors need to characterize when transfer works (e.g., by per-attribute analysis of part and pattern support, multiple random splits, and comparison to a no-sharing baseline) and to report variance, before the zero-shot capability can be accepted as a general property.
  2. [Sec. 4.1, Sec. 4.3.3] The benchmark is defined post hoc: back and tail attributes are excluded from the 204-attribute set because they are 'nosily labelled,' and the state-of-the-art comparison in Table 3 is restricted to the 32 attributes with more than 1000 positive samples. Both choices favor the proposed method, since sharing is less important and baselines are stronger on high-data attributes, while the noisy-label exclusion could remove exactly the attributes where the model might fail. The paper should either justify these exclusions with a principled, pre-specified criterion that is independent of the observed results, or show that the main conclusions are robust when all attributes (or all part groups) are included.
  3. [Sec. 4.3.3, Table 3] The comparison with PANDA and R-CNN is not clean: those methods use part annotations to train part detectors while CSN uses only image-level labels, which is a strength of CSN, but the reported numbers (65.5%* for PANDA vs. 65.2% for CSN) are accompanied by an unexplained asterisk and no statistical significance or error bars. The text says CSN obtains 'comparable performance,' but a reader cannot determine whether the 0.3-point gap is meaningful. Please explain the asterisk and report paired results or confidence intervals across multiple runs.
  4. [Sec. 4.3.1, Tables 1-2] The study of training sample size is limited to four attributes and two data sizes (200 and 500 images), with no repeated runs. The paper itself concedes in Sec. 4.3.1 that 'pattern features extracted at different location still varies' and that forced sharing can harm performance when per-attribute data are relatively large. Since the central advantage of CSN is supposed to grow as data shrink, the authors need a broader ablation over many attributes and several data sizes with error bars to actually demonstrate that trend, and to identify conditions under which sharing helps versus hurts.
minor comments (6)
  1. [Sec. 4.1] The phrase 'nosily labelled' should be 'noisily labelled'.
  2. [Fig. 2 caption] The caption appears garbled: 'unseen attribute wing redforwardunseen attribute forwardsupervision' should be split into readable text.
  3. [Tables 3 and 4 captions] The sentence 'The numbers are shown only for the 32 attributes that contain more than 1000 positive samples in this dataset. If we perform such comparison for attributes with smaller training set, the baseline always produces very poor result.' appears in both table captions, suggesting a copy-paste error, and the asterisk after 65.5% in Table 3 is never explained.
  4. [Sec. 4.3.4] In the soft-sharing variant, the attention map in Eq. (8) is a weighted sum of per-part attention maps, but the paper does not specify whether the combination is normalized or how the softmax over the learnable weight vector is applied after initialization; please clarify.
  5. [Sec. 4.3.5] The sentence 'we randomly select 20 attributes' gives no random seed or split identifier; please provide the actual split or run multiple splits and report the mean and standard deviation.
  6. [Sec. 4.4] The CelebA evaluation says it follows the protocol in [18], but the paper does not describe that protocol; specify the training/test split and which attributes are grouped into shared localization modules.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the zero-shot capability is a genuine forward test of composing learned part-location and appearance-pattern modules, and the paper's self-citations are not load-bearing.

full rationale

The paper's central derivation is not circular. In Sec. 3.3, the claim that an unseen attribute P_{i,j} can be recognized follows from the architectural definition in Eq. 4: the classifier is f(x; Theta, V_i, W_j^T) = softmax(W_j^T \bar{Q}'_i). Given this definition, the recognition model is indeed determined by the shared location module V_i and the shared pattern module W_j, and those modules can be learned from other attributes without any samples of P_{i,j}. This is a compositional/transferability claim, not a fitting of the target quantity: the held-out experiment in Sec. 4.3.5 randomly selects 20 unseen attributes, trains on the remaining 184, and evaluates on the unseen labels without fitting any parameter to those labels. The paper's own concession that 'pattern features extracted at different location still varies' and the poor zero-shot result for 'bill yellow' in Table 8 bear on the validity and generality of the transfer assumption, which is a correctness risk, not circularity. The two self-citations by the first author ([67], [68]) appear only in general related-work enumerations and are not used to justify the central premise or to forbid alternatives. No fitted input is renamed as a prediction, and no uniqueness theorem or ansatz is imported solely from the authors' prior work. The derivation chain is self-contained in the sense that the mechanism is specified by the paper's own equations and then tested against held-out data.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The central mechanism rests on inputs the paper does not derive: attributes decompose into transferable part and pattern concepts (a domain assumption the paper notes fails when data is abundant), image-level labels suffice to train attention localization without part annotations, the standard math of spatial softmax attention, and a post-hoc exclusion of two attribute families from the benchmark. The hand-chosen sharing assignment carries much of the benefit: initializing soft-sharing weights without the part prior (CSN-soft-1) drops AP from 65.1% to 63.8%, close to the 63.1% no-share baseline.

free parameters (4)
  • Part representation vectors Vi (one per part group) = learned; dimension equal to ResNext50 conv5 channels
    Ordinary learned parameters of the localization module (Eq. 1). Counted for completeness; they are trained by cross-entropy, not fit to the test claim.
  • Pattern classification weights Wj (one per pattern group) = learned; R^{d x 2} per pattern
    Ordinary learned binary-classifier weights (Eq. 4). Not ad hoc constants.
  • Attribute-to-module sharing assignment = hand-chosen from CUB/CelebA label names
    The data-expansion benefit comes from this grouping. Removing the prior, via all-ones soft-sharing initialization, drops AP from 65.1% to 63.8% (Tab. 4).
  • Training hyperparameters = LR 1e-4, decay 0.1 at epoch 50, 100 epochs, crop 446 from 512
    Disclosed in Sec. 4.2; no sensitivity analysis, so their influence is uncharacterized.
assumptions (4)
  • standard math Spatial softmax over inner-product attention maps provides differentiable, trainable localization
    Eqs. 1-3 define the attention mechanism; correctness follows from standard backpropagation.
  • domain assumption Each attribute Pi,j is the conjunction of an independent part i and pattern j, and Wj learned from other parts transfers to part i
    Secs. 3.2-3.3. The paper concedes this invariance is imperfect: sharing patterns 'will harm the performance' with large data (Sec. 4.3.1).
  • domain assumption Image-level attribute labels suffice to train part localization via attention, without part annotations
    Sec. 3.1.1 and Fig. 4; localization emerges from the classification losses in Eq. 5.
  • ad hoc to paper The back and tail attribute labels are noisy and can be excluded from the benchmark
    Sec. 4.1: 'we observe that the back and tail attributes are nosily labelled. Therefore, we exclude these labels and conduct experiments on the remaining 204 attributes.'

how reviews work

0 comments
Cite this review

Pith. "Pith review of Recognizing Part Attributes with Insufficient Data." pith.science (2026). https://pith.science/paper/GD2CJ3IY

@misc{pith2026190803335,
  author       = {Pith},
  title        = {Pith review of: Recognizing Part Attributes with Insufficient Data},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GD2CJ3IY}},
  note         = {Machine review of arXiv:1908.03335}
}
read the original abstract

Recognizing attributes of objects and their parts is important to many computer vision applications. Although great progress has been made to apply object-level recognition, recognizing the attributes of parts remains less applicable since the training data for part attributes recognition is usually scarce especially for internet-scale applications. Furthermore, most existing part attribute recognition methods rely on the part annotation which is more expensive to obtain. To solve the data insufficiency problem and get rid of dependence on the part annotation, we introduce a novel Concept Sharing Network (CSN) for part attribute recognition. A great advantage of CSN is its capability of recognizing the part attribute (a combination of part location and appearance pattern) that has insufficient or zero training data, by learning the part location and appearance pattern respectively from the training data that usually mix them in a single label. Extensive experiments on CUB-200-2011 [51], CelebA [35] and a newly proposed human attribute dataset demonstrate the effectiveness of CSN and its advantages over other methods, especially for the attributes with few training samples. Further experiments show that CSN can also perform zero-shot part attribute recognition. The code will be made available at https://github.com/Zhaoxiangyun/Concept-Sharing-Network.

Figures

Figures reproduced from arXiv: 1908.03335 by the authors.

Figure 1
Figure 1. In many datasets and real applications, the labeling of [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Overview of the training. Training images of differ￾ent attributes are forwarded through the CNN to obtain the image representation, then attribute samples with the same part are for￾warded through the same localization module and attribute sam￾ples with the same appearance pattern are forwarded through the same pattern recognition module. New attribute with no training data could be recognized as the combination of… view at source ↗
Figure 3
Figure 3. Overview of the inference. In the inference, given an image and recognizing attribute Pi,j , the image is forwarded through the corresponding location module i and pattern module j to obtain the final prediction. soft-max layer to produce final probability of an attribute. In contrast, our method learns an attention map for each attribute to weight the feature map so that the following ap￾pearance pattern could be l… view at source ↗
Figures from the paper (8 more)
Figure 4
Figure 4. Figure 4: Comparison of heat map generated by CSN w/o sharing [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: CSN performance gain against baseline on joint training of 204 attributes. The baseline method refers to a multi task learning [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: Qualitative results of CSN on CUB-200-2011 test set. For each pair of images in the figure, the left shows the input image. [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 7
Figure 7. Figure 7: Exemplar images of our human dataset. The heat map [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 8
Figure 8. Figure 8: Exemplar images of our human dataset [PITH_FULL_IMAGE:figures/full_fig_p013_8.png]
Figure 9
Figure 9. Figure 9: Exemplar images of our human dataset [PITH_FULL_IMAGE:figures/full_fig_p014_9.png]
Figure 10
Figure 10. Figure 10: Exemplar images of the CUB bird dataset. [PITH_FULL_IMAGE:figures/full_fig_p015_10.png]
Figure 11
Figure 11. Figure 11: Exemplar images of the CUB bird dataset. [PITH_FULL_IMAGE:figures/full_fig_p016_11.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

70 extracted references · 59 canonical work pages

  1. [1]

    How to transfer? zero-shot object recognition via hierarchical transfer of se- mantic attributes

    Ziad Al-Halah and Rainer Stiefelhagen. How to transfer? zero-shot object recognition via hierarchical transfer of se- mantic attributes. In WACV, pages 837–843. IEEE, 2015. 2

    work page 2015

  2. [2]

    Neural module networks

    Jacob Andreas, Marcus Rohrbach, Trevor Darrell, and Dan Klein. Neural module networks. In CVPR, pages 39–48,

    work page

  3. [3]

    Vqa: Visual question answering

    Stanislaw Antol, Aishwarya Agrawal, Jiasen Lu, Margaret Mitchell, Dhruv Batra, C Lawrence Zitnick, and Devi Parikh. Vqa: Visual question answering. InICCV, pages 2425–2433,

  4. [4]

    Mul- tiple object recognition with visual attention

    Jimmy Ba, V olodymyr Mnih, and Koray Kavukcuoglu. Mul- tiple object recognition with visual attention. arXiv preprint arXiv:1412.7755, 2014. 3

    arXiv 2014

  5. [5]

    De- scribing people: A poselet-based approach to attribute clas- sification

    Lubomir Bourdev, Subhransu Maji, and Jitendra Malik. De- scribing people: A poselet-based approach to attribute clas- sification. In ICCV, pages 1543–1550. IEEE, 2011. 2, 3

    work page 2011

  6. [6]

    Poselets: Body part detectors trained using 3d human pose annotations

    Lubomir Bourdev and Jitendra Malik. Poselets: Body part detectors trained using 3d human pose annotations. InICCV, pages 1365–1372. IEEE, 2009. 6

    work page 2009

  7. [7]

    An empirical study and analysis of generalized zero- shot learning for object recognition in the wild

    Wei-Lun Chao, Soravit Changpinyo, Boqing Gong, and Fei Sha. An empirical study and analysis of generalized zero- shot learning for object recognition in the wild. In ECCV, pages 52–68. Springer, 2016. 2

    work page 2016

  8. [8]

    Abc-cnn: An attention based convolutional neural network for visual question answering

    Kan Chen, Jiang Wang, Liang-Chieh Chen, Haoyuan Gao, Wei Xu, and Ram Nevatia. Abc-cnn: An attention based convolutional neural network for visual question answering. arXiv preprint arXiv:1511.05960, 2015. 3

    arXiv 2015

Show all 70 references
  1. [9]

    Pedestrian attribute recognition at far distance

    Yubin Deng, Ping Luo, Chen Change Loy, and Xiaoou Tang. Pedestrian attribute recognition at far distance. In Proceed- ings of the 22nd ACM international conference on Multime- dia, pages 789–792. ACM, 2014. 3

    2014

  2. [10]

    Attribute-centric recognition for cross-category generalization

    Ali Farhadi, Ian Endres, and Derek Hoiem. Attribute-centric recognition for cross-category generalization. In CVPR, pages 2352–2359. IEEE, 2010. 2, 3

    2010

  3. [11]

    Describing objects by their attributes

    Ali Farhadi, Ian Endres, Derek Hoiem, and David Forsyth. Describing objects by their attributes. InCVPR, pages 1778–

  4. [12]

    Learning visual at- tributes

    Vittorio Ferrari and Andrew Zisserman. Learning visual at- tributes. In NIPS, pages 433–440, 2008. 2

    2008

  5. [13]

    Model- agnostic meta-learning for fast adaptation of deep networks

    Chelsea Finn, Pieter Abbeel, and Sergey Levine. Model- agnostic meta-learning for fast adaptation of deep networks. In Proceedings of the 34th International Conference on Ma- chine Learning-Volume 70 , pages 1126–1135. JMLR. org,

  6. [14]

    Fine-grained recognition in the wild: A multi-task domain adaptation ap- proach

    Timnit Gebru, Judy Hoffman, and Li Fei-Fei. Fine-grained recognition in the wild: A multi-task domain adaptation ap- proach. In Proceedings of the IEEE International Confer- ence on Computer Vision, pages 1349–1358, 2017. 2

    2017

  7. [15]

    Fast r-cnn

    Ross Girshick. Fast r-cnn. In ICCV, pages 1440–1448, 2015. 1

    2015

  8. [16]

    Region-based convolutional networks for accurate object detection and segmentation

    Ross Girshick, Jeff Donahue, Trevor Darrell, and Jitendra Malik. Region-based convolutional networks for accurate object detection and segmentation. IEEE transactions on pattern analysis and machine intelligence , 38(1):142–158,

  9. [17]

    Evaluating ap- pearance models for recognition, reacquisition, and track- ing

    Douglas Gray, Shane Brennan, and Hai Tao. Evaluating ap- pearance models for recognition, reacquisition, and track- ing. In Proc. IEEE International Workshop on Performance Evaluation for Tracking and Surveillance (PETS), volume 3, pages 1–7. Citeseer, 2007. 3

    2007

  10. [18]

    Attributes for improved attributes: A multi-task network utilizing implicit and ex- plicit relationships for facial attribute classification

    Emily M Hand and Rama Chellappa. Attributes for improved attributes: A multi-task network utilizing implicit and ex- plicit relationships for facial attribute classification. InAAAI,

  11. [19]

    A simple general approach to inference about the tail of a distribution.The annals of statistics, pages 1163– 1174, 1975

    Bruce M Hill. A simple general approach to inference about the tail of a distribution.The annals of statistics, pages 1163– 1174, 1975. 2

    1975

  12. [20]

    Labeled faces in the wild: A database forstudying face recognition in unconstrained environments

    Gary B Huang, Marwan Mattar, Tamara Berg, and Eric Learned-Miller. Labeled faces in the wild: A database forstudying face recognition in unconstrained environments. In Workshop on faces in’Real-Life’Images: detection, align- ment, and recognition, 2008. 3

    2008

  13. [21]

    Spatial transformer networks

    Max Jaderberg, Karen Simonyan, Andrew Zisserman, et al. Spatial transformer networks. In NIPS, pages 2017–2025,

    2017

  14. [22]

    Human attribute recognition by rich appearance dictionary

    Jungseock Joo, Shuo Wang, and Song-Chun Zhu. Human attribute recognition by rich appearance dictionary. InCVPR, pages 721–728, 2013. 2

    2013

  15. [23]

    Adam: A method for stochastic optimization

    Diederik P Kingma and Jimmy Ba. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980 ,

    arXiv

  16. [24]

    Imagenet classification with deep convolutional neural net- works

    Alex Krizhevsky, Ilya Sutskever, and Geoffrey E Hinton. Imagenet classification with deep convolutional neural net- works. In NIPS, pages 1097–1105, 2012. 1

    2012

  17. [25]

    Describable visual attributes for face verifica- tion and image search

    Neeraj Kumar, Alexander Berg, Peter N Belhumeur, and Shree Nayar. Describable visual attributes for face verifica- tion and image search. IEEE Transactions on Pattern Anal- ysis and Machine Intelligence, 33(10):1962–1977, 2011. 2, 3

    1962

  18. [26]

    Attribute and simile classifiers for face veri- fication

    Neeraj Kumar, Alexander C Berg, Peter N Belhumeur, and Shree K Nayar. Attribute and simile classifiers for face veri- fication. In ICCV, pages 365–372. IEEE, 2009. 2, 3

    2009

  19. [27]

    Learning to detect unseen object classes by between- class attribute transfer

    Christoph H Lampert, Hannes Nickisch, and Stefan Harmel- ing. Learning to detect unseen object classes by between- class attribute transfer. In CVPR, pages 951–958. IEEE,

  20. [28]

    Attribute-based classification for zero-shot visual object categorization

    Christoph H Lampert, Hannes Nickisch, and Stefan Harmel- ing. Attribute-based classification for zero-shot visual object categorization. IEEE Transactions on Pattern Analysis and Machine Intelligence, 36(3):453–465, 2014. 2, 3

    2014

  21. [29]

    A richly annotated dataset for pedestrian at- tribute recognition

    Dangwei Li, Zhang Zhang, Xiaotang Chen, Haibin Ling, and Kaiqi Huang. A richly annotated dataset for pedestrian at- tribute recognition. arXiv preprint arXiv:1603.07054, 2016. 3

    2016 arXiv

  22. [30]

    Human attribute recognition by deep hierarchical con- texts

    Yining Li, Chen Huang, Chen Change Loy, and Xiaoou Tang. Human attribute recognition by deep hierarchical con- texts. In ECCV, pages 684–700. Springer, 2016. 1, 2, 3

    2016

  23. [31]

    Zero-shot recognition using dual visual- semantic mapping paths

    Yanan Li, Donghui Wang, Huanhang Hu, Yuetan Lin, and Yueting Zhuang. Zero-shot recognition using dual visual- semantic mapping paths. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR) , pages 5207–5215,

  24. [32]

    Localizing by describing: Attribute-guided attention lo- calization for fine-grained recognition

    Xiao Liu, Jiang Wang, Shilei Wen, Errui Ding, and Yuanqing Lin. Localizing by describing: Attribute-guided attention lo- calization for fine-grained recognition. In Thirty-First AAAI Conference on Artificial Intelligence, 2017. 2

    2017

  25. [33]

    Fully convolutional attention networks for fine-grained recognition

    Xiao Liu, Tian Xia, Jiang Wang, Yi Yang, Feng Zhou, and Yuanqing Lin. Fully convolutional attention networks for fine-grained recognition. arXiv preprint arXiv:1603.06765,

    arXiv

  26. [34]

    Deepfashion: Powering robust clothes recognition and retrieval with rich annotations

    Ziwei Liu, Ping Luo, Shi Qiu, Xiaogang Wang, and Xiaoou Tang. Deepfashion: Powering robust clothes recognition and retrieval with rich annotations. In CVPR, pages 1096–1104,

  27. [35]

    Deep learning face attributes in the wild

    Ziwei Liu, Ping Luo, Xiaogang Wang, and Xiaoou Tang. Deep learning face attributes in the wild. In ICCV, pages 3730–3738, 2015. 1, 3, 5, 7

    2015

  28. [36]

    Fully-adaptive feature shar- ing in multi-task networks with applications in person at- tribute classification

    Yongxi Lu, Abhishek Kumar, Shuangfei Zhai, Yu Cheng, Tara Javidi, and Rogerio Feris. Fully-adaptive feature shar- ing in multi-task networks with applications in person at- tribute classification. In CVPR, pages 5334–5343, 2017. 8

    2017

  29. [37]

    Transparency by design: Closing the gap be- tween performance and interpretability in visual reasoning

    David Mascharka, Philip Tran, Ryan Soklaski, and Arjun Majumdar. Transparency by design: Closing the gap be- tween performance and interpretability in visual reasoning. In CVPR, pages 4942–4950, 2018. 3

    2018

  30. [38]

    Re- current models of visual attention

    V olodymyr Mnih, Nicolas Heess, Alex Graves, et al. Re- current models of visual attention. In Advances in neural information processing systems, pages 2204–2212, 2014. 3

    2014

  31. [39]

    Zero-shot learning with semantic output codes

    Mark Palatucci, Dean Pomerleau, Geoffrey E Hinton, and Tom M Mitchell. Zero-shot learning with semantic output codes. In NIPS, pages 1410–1418, 2009. 2

    2009

  32. [40]

    Sun attribute database: Discovering, annotating, and recognizing scene attributes

    Genevieve Patterson and James Hays. Sun attribute database: Discovering, annotating, and recognizing scene attributes. In CVPR, pages 2751–2758. IEEE, 2012. 3

    2012

  33. [41]

    Coco attributes: At- tributes for people, animals, and objects

    Genevieve Patterson and James Hays. Coco attributes: At- tributes for people, animals, and objects. In ECCV, pages 85–100. Springer, 2016. 3

    2016

  34. [42]

    Optimization as a model for few-shot learning

    Sachin Ravi and Hugo Larochelle. Optimization as a model for few-shot learning. 2016. 2

    2016

  35. [43]

    You only look once: Unified, real-time object de- tection

    Joseph Redmon, Santosh Divvala, Ross Girshick, and Ali Farhadi. You only look once: Unified, real-time object de- tection. In CVPR, pages 779–788, 2016. 1

    2016

  36. [44]

    Faster r-cnn: Towards real-time object detection with region proposal networks

    Shaoqing Ren, Kaiming He, Ross Girshick, and Jian Sun. Faster r-cnn: Towards real-time object detection with region proposal networks. In NIPS, pages 91–99, 2015. 1

    2015

  37. [45]

    An embarrass- ingly simple approach to zero-shot learning

    Bernardino Romera-Paredes and Philip Torr. An embarrass- ingly simple approach to zero-shot learning. In ICML, pages 2152–2161, 2015. 2

    2015

  38. [46]

    Attribute learning in large-scale datasets

    Olga Russakovsky and Li Fei-Fei. Attribute learning in large-scale datasets. In ECCV, pages 1–14. Springer, 2010. 3

    2010

  39. [47]

    At- tention for fine-grained categorization

    Pierre Sermanet, Andrea Frome, and Esteban Real. At- tention for fine-grained categorization. arXiv preprint arXiv:1412.7054, 2014. 3

    2014 arXiv

  40. [48]

    Learning discriminative spatial representation for image classification

    Gaurav Sharma and Frederic Jurie. Learning discriminative spatial representation for image classification. In BMVC, pages 1–11. BMV A Press, 2011. 3

    2011

  41. [49]

    Per- son attribute recognition with a jointly-trained holistic cnn model

    Patrick Sudowe, Hannah Spitzer, and Bastian Leibe. Per- son attribute recognition with a jointly-trained holistic cnn model. In ICCV Workshops, pages 87–95, 2015. 3

    2015

  42. [50]

    Is learning the n-th thing any easier than learning the first? In NIPS, pages 640–646, 1996

    Sebastian Thrun. Is learning the n-th thing any easier than learning the first? In NIPS, pages 640–646, 1996. 2

    1996

  43. [51]

    The caltech-ucsd birds-200-2011 dataset

    Catherine Wah, Steve Branson, Peter Welinder, Pietro Per- ona, and Serge Belongie. The caltech-ucsd birds-200-2011 dataset. 2011. 1, 3, 5

    2011

  44. [52]

    Relational knowledge transfer for zero-shot learning

    Donghui Wang, Yanan Li, Yuetan Lin, and Yueting Zhuang. Relational knowledge transfer for zero-shot learning. In AAAI, volume 2, page 7, 2016. 2

    2016

  45. [53]

    Residual attention network for image classification

    Fei Wang, Mengqing Jiang, Chen Qian, Shuo Yang, Cheng Li, Honggang Zhang, Xiaogang Wang, and Xiaoou Tang. Residual attention network for image classification. In CVPR, pages 3156–3164, 2017. 3

    2017

  46. [54]

    Learn- ing models for object recognition from natural language de- scriptions

    Josiah Wang, Katja Markert, and Mark Everingham. Learn- ing models for object recognition from natural language de- scriptions. In BMVC, volume 1, page 2, 2009. 3

    2009

  47. [55]

    At- tribute recognition by joint recurrent learning of context and correlation

    Jingya Wang, Xiatian Zhu, Shaogang Gong, and Wei Li. At- tribute recognition by joint recurrent learning of context and correlation. In ICCV, pages 531–540, 2017. 2

    2017

  48. [56]

    Non-local neural networks

    Xiaolong Wang, Ross Girshick, Abhinav Gupta, and Kaim- ing He. Non-local neural networks. In CVPR, 2018. 3

    2018

  49. [57]

    Cbam: Convolutional block attention module

    Sanghyun Woo, Jongchan Park, Joon-Young Lee, and In So Kweon. Cbam: Convolutional block attention module. In ECCV, September 2018. 2

    2018

  50. [58]

    Zero- shot learning-the good, the bad and the ugly

    Yongqin Xian, Bernt Schiele, and Zeynep Akata. Zero- shot learning-the good, the bad and the ugly. arXiv preprint arXiv:1703.04394, 2017. 2

    2017 arXiv

  51. [59]

    Aggregated residual transformations for deep neural networks

    Saining Xie, Ross Girshick, Piotr Doll ´ar, Zhuowen Tu, and Kaiming He. Aggregated residual transformations for deep neural networks. In Computer Vision and Pattern Recogni- tion (CVPR), 2017 IEEE Conference on , pages 5987–5995. IEEE, 2017. 5

    2017

  52. [60]

    Show, attend and tell: Neural image caption gen- eration with visual attention

    Kelvin Xu, Jimmy Ba, Ryan Kiros, Kyunghyun Cho, Aaron Courville, Ruslan Salakhudinov, Rich Zemel, and Yoshua Bengio. Show, attend and tell: Neural image caption gen- eration with visual attention. In ICML, pages 2048–2057,

  53. [61]

    A large-scale car dataset for fine-grained categorization and verification

    Linjie Yang, Ping Luo, Chen Change Loy, and Xiaoou Tang. A large-scale car dataset for fine-grained categorization and verification. In CVPR, pages 3973–3981, 2015. 3

    2015

  54. [62]

    Part-based r-cnns for fine-grained category detection

    Ning Zhang, Jeff Donahue, Ross Girshick, and Trevor Dar- rell. Part-based r-cnns for fine-grained category detection. In ECCV, pages 834–849. Springer, 2014. 1, 2, 6

    2014

  55. [63]

    Deformable part descriptors for fine-grained recognition and attribute prediction

    Ning Zhang, Ryan Farrell, Forrest Iandola, and Trevor Dar- rell. Deformable part descriptors for fine-grained recognition and attribute prediction. In CVPR, pages 729–736, 2013. 2

    2013

  56. [64]

    Panda: Pose aligned net- works for deep attribute modeling

    Ning Zhang, Manohar Paluri, Marc’Aurelio Ranzato, Trevor Darrell, and Lubomir Bourdev. Panda: Pose aligned net- works for deep attribute modeling. In CVPR, pages 1637– 1644, 2014. 1, 2, 3, 6

    2014

  57. [65]

    Zero-shot recog- nition via structured prediction

    Ziming Zhang and Venkatesh Saligrama. Zero-shot recog- nition via structured prediction. In ECCV, pages 533–548. Springer, 2016. 2

    2016

  58. [66]

    A large-scale attribute dataset for zero-shot learning

    Bo Zhao, Yanwei Fu, Rui Liang, Jiahong Wu, Yonggang Wang, and Yizhou Wang. A large-scale attribute dataset for zero-shot learning. arXiv preprint arXiv:1804.04314, 2018. 3

    2018 arXiv

  59. [67]

    A modulation module for multi-task learning with applications in image retrieval

    Xiangyun Zhao, Haoxiang Li, Xiaohui Shen, Xiaodan Liang, and Ying Wu. A modulation module for multi-task learning with applications in image retrieval. In Proceedings of the European Conference on Computer Vision (ECCV) , pages 401–416, 2018. 2

    2018

  60. [68]

    Pseudo mask augmented object detection

    Xiangyun Zhao, Shuang Liang, and Yichen Wei. Pseudo mask augmented object detection. In Proceedings of the IEEE conference on computer vision and pattern recogni- tion, pages 4061–4070, 2018. 1

    2018

  61. [69]

    Learning deep features for discrimi- native localization

    Bolei Zhou, Aditya Khosla, Agata Lapedriza, Aude Oliva, and Antonio Torralba. Learning deep features for discrimi- native localization. In CVPR, pages 2921–2929, 2016. 2, 3, 6

    2016

  62. [70]

    Pedestrian attribute classification in surveillance: Database and evaluation

    Jianqing Zhu, Shengcai Liao, Zhen Lei, Dong Yi, and Stan Li. Pedestrian attribute classification in surveillance: Database and evaluation. In CVPR Workshops, pages 331– 338, 2013. 3 A. Experimental Attributes In Figure 5, we mentioned that we performed experiments on 204 attrib...

    2013

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.