Pith. sign in

REVIEW 5 major objections 5 minor 57 references

Text-guided Zero-Shot Object Localization

T0 review · 5 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read The paper claims that text-guided zero-shot object localization works by aligning prompt-word-enhanced CLIP text embeddings with image patches, producing density maps whose peaks are object centers without labeled examples.

desk verdict An engineering extension of zero-shot counting with an under-supported TSSM module; the paper's own ablation shows counting metrics degrading, so the central claim needs much better evidence. read the letter →

arxiv 2411.11357 v1 pith:24WZCL3Y submitted 2024-11-18 cs.CV

Jingjing Wang , Xinglin Piao , Zongzhi Gao , Bo Li , Yong Zhang , Baocai Yin This is my paper · ORCID
classification cs.CV
keywords zero-shotobjectlocalizationtext-guidedCLIPtextself-similaritymatchingdensitymapestimationfew-shotcross-modallearningFSC-147
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 prove that object localization can be done with zero labeled examples, using only a text prompt describing the target object. It builds a framework around a frozen CLIP model, which already maps images and text into a shared space, and adds a text self-similarity matching (TSSM) module that weights text features toward the prompt word. The authors report that the resulting ZSOLNet locates objects on FSC-147, CARPK, and ShanghaiTech with accuracy comparable to recent few-shot localization methods, and sometimes above fully supervised ones in dense scenes. If the result holds, it would let users localize arbitrary categories by typing a phrase, removing the annotation bottleneck.

What carries the argument

The central mechanism is the Text Self-Similarity Matching (TSSM) module, which takes the sentence-level text embedding $\varepsilon_t$ and the prompt-word embedding $\varepsilon_o$, computes their cosine similarity $W$, and forms the self-supporting embedding $\varepsilon_{ts} = W \cdot \varepsilon_t + \varepsilon_o$. This embedding then replaces the original text feature in the CLIP alignment. The other machinery is a two-stage training loop: first a contrastive loss aligns image patches near density peaks with the text embedding, then an MSE loss refines the predicted density map; post-processing thresholds $\alpha$ and $\beta$ extract final object centers.

What would settle it

Run the same ZSOL pipeline with and without TSSM on multiple seeds and datasets, then compare F1 distributions at both thresholds; if the gap is not consistently positive or overlaps zero, the module's claimed contribution fails. A simpler probe: replace Eq. (5) with concatenation or simple averaging of the two embeddings; if localization accuracy does not drop, the similarity-weighting mechanism is not doing the claimed work.

Watch

Extended reading notes

Core claim

This paper claims the first zero-shot object localization framework driven purely by prompt words: given an image and a text phrase, the model outputs a density map, and local maxima mark object centers. The central discovery is that re-weighting the CLIP text embedding by its cosine similarity to the prompt-word embedding (the TSSM module) substantially improves localization precision, and that a frozen CLIP backbone can be adapted to dense localization through contrastive patch-level pretraining followed by MSE training. Reported results on FSC-147, CARPK, and ShanghaiTech place the zero-shot model within the range of few-shot and fully supervised methods, and on ShanghaiTechA it exceeds several supervised baselines on F1 and AP.

Load-bearing premise

The TSSM module's core assumption is that adding the prompt-word embedding weighted by cosine similarity to the sentence embedding makes the text representation sharper for localization; this is an empirical heuristic, tested on one dataset without statistical significance checks.

Editorial extensions

If this is right

  • Users could localize any object category by typing a phrase, with no per-category training data or annotation effort.
  • The zero-shot model is competitive in dense scenes, matching or exceeding few-shot baselines on ShanghaiTechA and CARPK while staying close on FSC-147.
  • Removing the TSSM module drops F1 substantially on FSC-147, so the text re-weighting step is a necessary part of the reported accuracy.
  • Because the CLIP encoders stay frozen, the same trained decoder transfers to new datasets and categories without retraining.

Reading between the lines

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

  • An untested extension is to read the density map's integral as a count, turning the same model into a zero-shot counter; this would unify two tasks the paper treats separately.
  • The 'first zero-shot localization' claim is narrower than it sounds: earlier text-guided zero-shot counting methods already used CLIP-style prompts, so the new contribution is applying the idea to center-point localization, not inventing cross-modal prompting.
  • A stress test the paper does not report is varying the prompt wording (synonyms, plurals, or captions instead of noun phrases); TSSM's cosine weighting may be sensitive to phrasing, which would matter in practice.
  • The method inherits CLIP's training distribution, so its zero-shot scope is bounded by the concepts CLIP has seen; testing on genuinely novel object categories would reveal the practical boundary.
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

5 major / 5 minor

Summary. The paper proposes Zero-Shot Object Localization (ZSOL), a framework that uses CLIP-based visual and textual encoders together with a Text Self-Similarity Matching (TSSM) module to localize objects from prompt words without class-specific labeled examples. The model is trained on FSC-147 with frozen encoders, a contrastive loss, and an MSE loss on predicted density maps, then post-processes density peaks with two thresholds. The authors report localization and counting results on FSC-147, ShanghaiTech A/B, and CARPK, comparing against fully supervised and few-shot methods, and conclude that ZSOL achieves performance comparable to few-shot approaches while requiring no annotated target-category data.

Significance. If the claims are substantiated, the paper would provide a useful first benchmark for text-guided zero-shot object localization, which is a plausible extension of recent zero-shot counting work. The use of frozen CLIP encoders, a text self-similarity module, and density-map post-processing is a reasonable design that could be a practical baseline. The paper includes comparisons across several standard datasets and reports both localization and counting metrics. However, the central novel component, TSSM, is not specified precisely enough to be reproduced, and its only supporting ablation is a single-run experiment with mixed and statistically unvalidated results. The per-dataset threshold tuning also weakens the zero-shot claim. These issues currently prevent the paper from being accepted as a reliable technical contribution.

major comments (5)
  1. [§III-B1, Eqs. (4)-(5)] The TSSM module is not specified precisely enough to be implemented. The dimensions of εt and εo are never defined, and the statement that "n = 3 represents three pairs of samples fed into each batch" is inconsistent with n being used as the summation index in Eq. (4), where n should be the embedding dimension. The convolution mentioned in the text ("The title embedding obtained by the convolution is again computed by cosine similarity") is never written out: the kernel, stride, padding, and input/output shapes are all absent. Because TSSM is the paper's main novelty, these missing definitions are load-bearing.
  2. [Table V (ablation study)] The evidence that TSSM improves localization is not convincing. The ablation shows F1/AP/AR gains with TSSM, but MAE is consistently worse with the module (28.46 vs. 24.33 on validation, 20.78 vs. 18.21 on test), while test MSE is essentially unchanged (100.14 vs. 105.43). This pattern suggests TSSM may sharpen density peaks rather than improve density-map accuracy. The ablation is performed on one dataset, without multiple seeds, error bars, or significance tests. The claim that "TSSM has a positive effect on improving zero-shot object localization performance" is therefore not statistically established.
  3. [§III-B2 and §III-B3] The training objectives and decoder architecture are not described precisely. The "contrastive loss" and "MSE loss" are only described in words; no equations, weighting coefficients, or exact definitions are given. The "joint embedding decoder" that maps multimodal features to a density map is not architecturally specified. The two-stage training procedure (20 contrastive epochs followed by 200 MSE epochs) is also underspecified with regard to batch size, image crop size, and learning-rate schedule. As a result, the experimental section is not reproducible from the manuscript alone.
  4. [§III-B3 (post-processing thresholds)] The peak thresholds α and β are selected per dataset and are not derived from a validation procedure. The text states that α = 5/255 is used for dense datasets and α = 10/255 for sparse datasets, and β = 0.06 globally. Because these thresholds are tuned on the test datasets and no sensitivity analysis is provided, it is unclear how much of the reported performance depends on this manual calibration. This weakens the zero-shot generalization claim, since per-dataset threshold tuning is a form of test-set adaptation.
  5. [§IV-D and Tables I-IV] The comparison with prior methods is not always apples-to-apples. For baselines without a double-threshold setting, the paper uses their reported SOTA results, while for others it uses thresholds σs and σl that may differ from the original papers; the exact provenance of each baseline number is not itemized. Additionally, the ZSOL model is trained on a cropped-image protocol and tested with a sliding window, which the paper itself notes can inflate test performance relative to validation. The tables show many missing entries for baselines, making it difficult to assess whether the reported gains are consistent across settings.
minor comments (5)
  1. [Eq. (2)] The notation Wit = f(Ie, Te) is introduced without explaining the projection layer f(·) or the dimensionality of Wit; please define all symbols in this equation.
  2. [§III-B1 and Fig. 4] The terms "title", "prompt word", and "sentence" are used interchangeably; clarify the relationship between the user-provided prompt, the expanded sentence, and the extracted title embedding.
  3. [Table V] The column layout of Table V is corrupted in the displayed text, with "MAE MSE√" on one line and the numeric values not clearly aligned; please reformat the table so that validation/test splits and metric columns are unambiguous.
  4. [§IV-D] There is a typo in the sentence "the ZSOL model exhibits excellent localization performance in scenarios where the object object localization is highly distinguished from the background"; remove the duplicated "object".
  5. [§V] The conclusion calls the framework "a high-performance benchmark" but no code, model weights, or evaluation protocol are released; please state whether these will be made available.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the TSSM benefit is empirically ablated and the main results are benchmarked against external methods.

full rationale

The paper's central claim is an empirical method: CLIP features plus a TSSM text-feature re-weighting are trained with MSE and contrastive losses on FSC-147 and evaluated on unseen categories and other datasets. Equations (4)-(5) define epsilon_ts as a cosine-weighted combination of sentence and prompt embeddings; the paper does not claim that this definition alone implies localization accuracy, and the benefit is tested in Table V as a with/without-TSSM ablation, not derived from the equations. The main tables compare against external fully-supervised and few-shot baselines (IIM, CLTR, FIDTM, BMNet+, CACL, FSOL, etc.), so the reported numbers are falsifiable outside the paper's own fitted values. The FSOL reference [39] and cell-localization reference [13] share authors with the present paper, but they are used as a baseline and related work, not as evidence that the TSSM construction is correct or as a uniqueness theorem. The per-dataset choice of post-processing thresholds alpha (5/255 vs 10/255) and beta is a standard peak-picking hyperparameter; it is not a parameter fitted to the localization metric and then renamed a prediction, and no equation reduces to itself. The lack of multiple seeds or error bars in Table V is a statistical robustness concern, not a circularity. Hence no significant circularity.

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

The method relies on pre-trained CLIP, a standard supervised training stage on FSC-147, and several hand-chosen thresholds. No new physical entities or exotic constructs are introduced.

free parameters (4)
  • alpha (peak threshold) = 5/255 for dense datasets, 10/255 for sparse datasets
    Chosen by hand per dataset to separate object peaks from background noise; directly affects precision and recall.
  • beta (global lower threshold) = 0.06
    Manually set to discard low-value candidate points; impacts false positive rate.
  • title embedding length limit = 3 tokens
    The prompt word mapping used as the convolution kernel is limited to length 3, a design choice not derived from data.
  • sliding window size and stride = 384x384, stride 128
    Used only at test time; influences the number of density maps and final localization accuracy.
assumptions (4)
  • domain assumption OpenAI CLIP ViT-B/16 provides a meaningful shared embedding space for images and text.
    The entire framework relies on CLIP features trained on web-scale data to align text prompts with image patches.
  • domain assumption Point annotations in the FSC-147 training set are valid supervision for learning a general density-map predictor.
    The model is trained with MSE loss against ground-truth density maps, assuming that knowledge transfers to unseen categories and other datasets.
  • domain assumption Local maxima in the predicted density map correspond to object centers.
    Post-processing uses max pooling and peak detection to extract object coordinates, which assumes the density map is a reliable indicator of object locations.
  • domain assumption The two-stage training schedule (contrast pretraining followed by MSE fine-tuning) converges to a useful solution.
    No convergence analysis is provided; the schedule is a heuristic choice.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Text-guided Zero-Shot Object Localization." pith.science (2026). https://pith.science/paper/24WZCL3Y

@misc{pith2026241111357,
  author       = {Pith},
  title        = {Pith review of: Text-guided Zero-Shot Object Localization},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/24WZCL3Y}},
  note         = {Machine review of arXiv:2411.11357}
}
read the original abstract

Object localization is a hot issue in computer vision area, which aims to identify and determine the precise location of specific objects from image or video. Most existing object localization methods heavily rely on extensive labeled data, which are costly to annotate and constrain their applicability. Therefore, we propose a new Zero-Shot Object Localization (ZSOL) framework for addressing the aforementioned challenges. In the proposed framework, we introduce the Contrastive Language Image Pre-training (CLIP) module which could integrate visual and linguistic information effectively. Furthermore, we design a Text Self-Similarity Matching (TSSM) module, which could improve the localization accuracy by enhancing the representation of text features extracted by CLIP module. Hence, the proposed framework can be guided by prompt words to identify and locate specific objects in an image in the absence of labeled samples. The results of extensive experiments demonstrate that the proposed method could improve the localization performance significantly and establishes an effective benchmark for further research.

Figures

Figures reproduced from arXiv: 2411.11357 by the authors.

Figure 1
Figure 1. Paradigm of object localization. (a) Fully Supervised: Costly due to [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Workflow for Zero-Shot Object Localization. (a) The encoder param [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. The overall ZSOL framework. (a) Text self-similarity matching. [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Schematic diagram of TSSM module. Through text self-similarity matching and feature weighting, the recognition capability of the ZSOL model for [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 5
Figure 5. Figure 5: Post-processing steps. By comprehensively applying pooling, filtering, and enhancement techniques to the coordinate points predicted by the density [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 6
Figure 6. Figure 6: The visualization results of the localization performance of the FSC [PITH_FULL_IMAGE:figures/full_fig_p009_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

57 extracted references · 52 canonical work pages

  1. [1]

    Rethinking classification and localization for object detection,

    Y . Wu, Y . Chen, L. Yuan, Z. Liu, L. Wang, H. Li, and Y . Fu, “Rethinking classification and localization for object detection,” in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020, pp. 10186–10195

    work page 2020

  2. [2]

    Weakly Supervised Object Localization and Detection: A Survey,

    D. Zhang, J. Han, G. Cheng, and M. H. Yang, “Weakly Supervised Object Localization and Detection: A Survey,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 44, no. 9, pp.5866–5885, 2022

    work page 2022

  3. [3]

    Locnet: Improving localization accuracy for object detection,

    S. Gidaris and N. Komodakis, “Locnet: Improving localization accuracy for object detection,” in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2016, pp. 789–798

    work page 2016

  4. [4]

    Weakly-supervised semantic segmentation by iteratively mining common object features,

    X. Wang, S. You, X. Li, and H. Ma, “Weakly-supervised semantic segmentation by iteratively mining common object features,” in: Pro- ceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018, pp. 1354–1362

    work page 2018

  5. [5]

    Object localization and border detection criteria design in edge-based image segmentation: automated learning from examples,

    M. Brejl and M. Sonka, “Object localization and border detection criteria design in edge-based image segmentation: automated learning from examples,” IEEE Transactions on Medical Imaging , vol. 19, no. 10, pp. 973–985, 2000

    work page 2000

  6. [6]

    Local All-Pair Correspondence for Point Tracking,

    S. Cho, J. Huang, J. Nam, H. An, S. Kim, and J. Y . Lee, “Local All-Pair Correspondence for Point Tracking,” arXiv preprint, arXiv:2407.15420, 2024

    arXiv 2024

  7. [7]

    UcoSLAM: Simultaneous localization and mapping by fusion of keypoints and squared planar markers,

    R. Muñoz-Salinas and R. Medina-Carnicer, “UcoSLAM: Simultaneous localization and mapping by fusion of keypoints and squared planar markers,” Pattern Recognition, vol. 101, p. 107193, 2020

    work page 2020

  8. [8]

    UA V-Satellite View Synthesis for Cross-View Geo-Localization,

    X. Tian, J. Shao, D. Ouyang, and H. T. Shen, “UA V-Satellite View Synthesis for Cross-View Geo-Localization,” IEEE Transactions on Circuits and Systems for Video Technology , vol. 32, no. 7, pp. 4804– 4815, 2022

    work page 2022

Show all 57 references
  1. [9]

    Gaussian yolov3: An accurate and fast object detector using localization uncertainty for autonomous driving,

    J. Choi, D. Chun, H. Kim, and H.-J. Lee, “Gaussian yolov3: An accurate and fast object detector using localization uncertainty for autonomous driving,” in: Proceedings of the IEEE/CVF International Conference on Computer Vision, 2019, pp. 502–511

    2019

  2. [10]

    Localization and navigation in autonomous driving: Threats and countermeasures,

    Q. Luo, Y . Cao, J. Liu, and A. Benslimane, “Localization and navigation in autonomous driving: Threats and countermeasures,” IEEE Wireless Communications, vol. 26, no. 4, pp. 38–45, 2019

    2019

  3. [11]

    Fusion- driven deep feature network for enhanced object detection and tracking in video surveillance systems,

    D. K. Jain, X. Zhao, C. Gan, P. K. Shukla, A. Jain, S. Sharma, "Fusion- driven deep feature network for enhanced object detection and tracking in video surveillance systems," Information Fusion, vol. 109, p. 102429, 2024

    2024

  4. [12]

    Automatic detection of mind wandering from video in the lab and in the classroom,

    N. Bosch and S. K. D’mello, “Automatic detection of mind wandering from video in the lab and in the classroom,” IEEE Transactions on Affective Computing, vol. 12, no. 4, pp. 974–988, 2019

    2019

  5. [13]

    Multi- scale hypergraph-based feature alignment network for cell localization,

    B. Li, Y . Zhang, C. Zhang, X. Piao, Y . Hu, and B. Yin, “Multi- scale hypergraph-based feature alignment network for cell localization,” Pattern Recognition, vol. 149, pp. 110260, 2024

    2024

  6. [14]

    Few-Shot Object Detection: A Survey,

    S. Antonelli, M. G. Cacace, G. Zappella, F. R. Lippiello, P. P. De Palma, and M. D. Perna, “Few-Shot Object Detection: A Survey,” ACM Computing Surveys, vol. 54, no. 11s, pp. 1–37, 2022

    2022

  7. [15]

    Generalized few-shot object detection without forgetting,

    Z. Fan, Y . Ma, Z. Li, and J. Sun, “Generalized few-shot object detection without forgetting,” in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2021, pp. 4527–4536

    2021

  8. [16]

    Semantic relation reasoning for shot-stable few-shot object detection,

    C. Zhu, F. Chen, U. Ahmed, Z. Shen, and M. Savvides, “Semantic relation reasoning for shot-stable few-shot object detection,” in: Pro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021, pp. 8782–8791

    2021

  9. [17]

    Learning to Detect Unseen Object Classes by Between-Class Attribute Transfer,

    C. H. Lampert, H. Nickisch, and S. Harmeling, “Learning to Detect Unseen Object Classes by Between-Class Attribute Transfer,” in: Pro- ceedings of IEEE Computer Vision and Pattern Recognition , 2009, pp. 951–958

    2009

  10. [18]

    Zero-Shot Learning Through Cross-Modal Transfer,

    R. Socher, M. Ganjoo, C. D. Manning, A. Ng, "Zero-Shot Learning Through Cross-Modal Transfer," in: Proceedings of Neural Information Processing Systems, 2013, pp. 935–943

    2013

  11. [20]

    Vlcounter: Text-Aware Visual Representation for Zero-Shot Object Counting,

    S. Kang, W. Moon, E. Kim, J. P. Heo, "Vlcounter: Text-Aware Visual Representation for Zero-Shot Object Counting," in Proceedings of the ACM International Conference on Multimedia , 2024, vol. 38, no. 3, pp. 2714–2722

    2024

  12. [21]

    Clip-Count: Towards Text-Guided Zero-Shot Object Counting,

    R. Jiang, L. Liu, and C. Chen, “Clip-Count: Towards Text-Guided Zero-Shot Object Counting,” in: Proceedings of the ACM International Conference on Multimedia , 2023, pp. 4535–4545

    2023

  13. [22]

    CountGD: Multi-Modal Open-World Counting,

    N. Amini-Naieni, T. Han, and A. Zisserman, “CountGD: Multi-Modal Open-World Counting,” arXiv preprint arXiv:2407.04619 , 2024

    2024 arXiv

  14. [23]

    Point Segment and Count: A Generalized Framework for Object Counting,

    Z. Huang, M. Dai, Y . Zhang, et al., “Point Segment and Count: A Generalized Framework for Object Counting,” in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2024, pp. 17067–17076

    2024

  15. [24]

    Metric learning with back- ground noise class for few-shot detection of rare sound events,

    K. Shimada, Y . Koyama, and A. Inoue, “Metric learning with back- ground noise class for few-shot detection of rare sound events,” in: Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing , 2020, pp. 616–620

    2020

  16. [25]

    Towards Zero-Shot Learning: A Brief Review and an Attention-Based Embedding Net- work,

    G. S. Xie, Z. Zhang, H. Xiong, L. Shao, X. Li, "Towards Zero-Shot Learning: A Brief Review and an Attention-Based Embedding Net- work," IEEE Transactions on Circuits and Systems for Video Technology, vol. 33, no. 3, pp. 1181–1197, 2022

    2022

  17. [26]

    Learning transferable visual models from natural language supervi- sion,

    A. Radford, J. W. Kim, C. Hallacy, A. Ramesh, G. Goh, S. Agarwal, G. Sastry, A. Askell, P. Mishkin, J. Clark, G. Krueger, and I. Sutskever, “Learning transferable visual models from natural language supervi- sion,” in: Proceedings of the International Conference on Machine Lea...

    2021

  18. [27]

    Self-support few-shot semantic segmentation,

    Q. Fan, W. Pei, Y . W. Tai, and C. K. Tang, “Self-support few-shot semantic segmentation,” in: Proceedings of the European Conference on Computer Vision , 2022, pp. 701–719

    2022

  19. [28]

    Point in, box out: Beyond count- ing persons in crowds,

    Y . Liu, M. Shi, Q. Zhao, and X. Wang, “Point in, box out: Beyond count- ing persons in crowds,” in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2019, pp. 6469–6478

    2019

  20. [29]

    Small instance detection by integer programming on object density maps,

    Z. Ma, L. Yu, and A. B. Chan, “Small instance detection by integer programming on object density maps,” in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2015, pp. 3689–3697

    2015

  21. [30]

    Dense point prediction: A simple baseline for crowd counting and localization,

    Y . Wang, X. Hou, and L.-P. Chau, “Dense point prediction: A simple baseline for crowd counting and localization,” in: Proceedings of the IEEE International Conference on Multimedia , 2021, pp. 1–6

    2021

  22. [31]

    Recurrent Attentive Zooming for Joint Crowd Counting and Precise Localization,

    C. Liu, X. Weng, and Y . Mu, “Recurrent Attentive Zooming for Joint Crowd Counting and Precise Localization,” in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2019, pp. 1217–1226

    2019

  23. [32]

    DeGPR: Deep Guided Posterior Regularization for Multi- Class Cell Detection and Counting,

    A. K. Tyagi, C. Mohapatra, P. Das, G. Makharia, L. Mehra, P. A. P., and Mausam, “DeGPR: Deep Guided Posterior Regularization for Multi- Class Cell Detection and Counting,” in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2023, pp. 23913–23923

    2023

  24. [33]

    Rethinking counting and localization in crowds: A purely JOURNAL OF LATEX CLASS FILES, VOL. XX, NO. X, NOV .2024 11 point-based framework,

    Q. Song, C. Wang, Z. Jiang, Y . Wang, Y . Tai, C. Wang, J. Li, F. Huang, and Y . Wu, “Rethinking counting and localization in crowds: A purely JOURNAL OF LATEX CLASS FILES, VOL. XX, NO. X, NOV .2024 11 point-based framework,” in: Proceedings of the IEEE/CVF International Confe...

    2024

  25. [34]

    Locally-Enriched Cross-Reconstruction for Few-Shot Fine-Grained Image Classification,

    X. Li, Q. Song, J. Wu, R. Zhu, Z. Ma, and J.-H. Xue, “Locally-Enriched Cross-Reconstruction for Few-Shot Fine-Grained Image Classification,” IEEE Transactions on Circuits and Systems for Video Technology , vol. 33, no. 12, pp. 7530–7540, 2023

    2023

  26. [35]

    Cross-Modal Contrastive Pre- training for Few-Shot Skeleton Action Recognition,

    M. Lu, S. Yang, X. Lu and J. Liu, “Cross-Modal Contrastive Pre- training for Few-Shot Skeleton Action Recognition,” IEEE Transactions on Circuits and Systems for Video Technology , vol. 34, no. 1, pp. 1–12, 2024

    2024

  27. [36]

    Learning to count everything,

    V . Ranjan, U. Sharma, T. Nguyen, and M. Hoai, “Learning to count everything,” in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2021, pp. 3394–3403

    2021

  28. [37]

    Few-Shot Object Counting and Detection,

    T. Nguyen, C. Pham, K. Nguyen, and M. Hoai, “Few-Shot Object Counting and Detection,” in: Proceedings of the European Conference on Computer Vision , vol. 13680, 2022, pp. 348–365

    2022

  29. [38]

    Few-shot object counting with similarity-aware feature enhancement,

    Z. You, K. Yang, W. Luo, X. Lu, L. Cui, and X. Le, “Few-shot object counting with similarity-aware feature enhancement,” in: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision , 2023, pp. 6315–6324

    2023

  30. [39]

    Few-shot Object Localization,

    Y . Ren, B. Li, C. Zhang, Y . Zhang, and B. Yin, “Few-shot Object Localization,” arXiv preprint, arXiv:2403.12466, 2024

    2024 arXiv

  31. [40]

    Dense center-direction regres- sion for object counting and localization with point supervision,

    D. Tabernik, J. Muhovi ˇc, and D. Sko ˇcaj, “Dense center-direction regres- sion for object counting and localization with point supervision,” Pattern Recognition, vol. 153, pp. 110540, 2024

    2024

  32. [41]

    Zero-Shot Action Recognition with Transformer-based Video Semantic Embedding,

    K. Doshi and Y . Yilmaz, “Zero-Shot Action Recognition with Transformer-based Video Semantic Embedding,” in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2023, pp. 4859–4868

    2023

  33. [42]

    Dynamic Momentum Adaptation for Zero-Shot Cross-Domain Crowd Counting,

    Q. Wu, J. Wan, and A. B. Chan, “Dynamic Momentum Adaptation for Zero-Shot Cross-Domain Crowd Counting,” in: Proceedings of the ACM International Conference on Multimedia , 2021, pp. 658–666

    2021

  34. [43]

    Focal Inverse Distance Trans- form Maps for Crowd Localization,

    D. Liang, W. Xu, Y . Zhu, and Y . Zhou, “Focal Inverse Distance Trans- form Maps for Crowd Localization,” IEEE Transactions on Multimedia, vol. 25, pp. 6040–6052, 2023

    2023

  35. [44]

    Segment Anything,

    A. Kirillov, E. Mintun, N. Ravi, H. Mao, C. Rolland, L. Gustafson, T. Xiao, S. Whitehead, A. C. Berg, W.-Y . Lo, P. Dollar, R. Girshick, “Segment Anything,” in: Proceedings of the IEEE/CVF International Conference on Computer Vision , 2023, pp. 4015–4026

    2023

  36. [45]

    Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection,

    S. Liu, Z. Zeng, T. Ren, F. Li, H. Zhang, J. Yang, Q. Jiang, C. Li, J. Yang, H. Su, J. Zhu, L. Zhang, “Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection,” arXiv preprint arXiv:2303.05499, 2023

    2023 arXiv

  37. [46]

    Learning To Count Everything,

    V . Ranjan, U. Sharma, T. Nguyen, and M. Hoai, “Learning To Count Everything,” in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2021, pp. 3394–3403

    2021

  38. [47]

    Represent, Compare, and Learn: A Similarity-Aware Framework for Class-Agnostic Counting,

    M. Shi, H. Lu, C. Feng, C. Liu, and Z. Cao, “Represent, Compare, and Learn: A Similarity-Aware Framework for Class-Agnostic Counting,” in: Proceedings of the 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022, pp. 9519–9528

    2022

  39. [48]

    Learning Independent In- stance Maps for Crowd Localization,

    J. Gao, T. Han, Q. Wang, Y . Yuan, and X. Li, “Learning Independent In- stance Maps for Crowd Localization,” arXiv preprint, arXiv:2012.04164, 2022

    2012 arXiv

  40. [49]

    Crowd Localization From Gaussian Mixture Scoped Knowledge and Scoped Teacher,

    J. Wang, J. Gao, Y . Yuan, and Q. Wang, “Crowd Localization From Gaussian Mixture Scoped Knowledge and Scoped Teacher,” IEEE Transactions on Image Processing , vol. 32, pp. 1802–1814, 2023

    2023

  41. [50]

    Counting and Locating Anything: Class-agnostic Few-shot Object Counting and Localization,

    Y . Zhang, H. Zhang, and Q. Zhao, “Counting and Locating Anything: Class-agnostic Few-shot Object Counting and Localization,” in: Pro- ceedings of the IEEE International Conference on Multimedia and Expo, 2023, pp. 2921–2926

    2023

  42. [51]

    STEERER: Resolving Scale Variations for Counting and Localization via Selective Inheritance Learning,

    T. Han, L. Bai, L. Liu, and W. Ouyang, “STEERER: Resolving Scale Variations for Counting and Localization via Selective Inheritance Learning,” in: Proceedings of the IEEE/CVF International Conference on Computer Vision , 2023, pp. 21848–21859

    2023

  43. [52]

    An End-to-End Transformer Model for Crowd Localization,

    D. Liang, W. Xu, and X. Bai, “An End-to-End Transformer Model for Crowd Localization,” in: Proceedings of the European Conference on Computer Vision, 2022, pp. 38–54

    2022

  44. [53]

    Dynamic Example Network for Class Agnostic Counting and Localization,

    X. Yan Liu, G. Li, Y . Qi, Z. Yan, W. Zhang, and Q. Huang, “Dynamic Example Network for Class Agnostic Counting and Localization,” available at SSRN: https://ssrn.com/abstract=4882210 or http://dx.doi. org/10.2139/ssrn.4882210, 2024

    doi

  45. [54]

    Locate, Size, and Count: Accurately Resolving People in Dense Crowds via Detection,

    D. B. Sam, S. V . Peri, M. N. Sundararaman, A. Kamath, and R. V . Babu, “Locate, Size, and Count: Accurately Resolving People in Dense Crowds via Detection,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 43, no. 8, pp. 2739–2751, 2021

    2021

  46. [55]

    A Low-Shot Object Counting Network With Iterative Prototype Adaptation,

    N. Ðuki ´c, A. Lukežiˇc, V . Zavrtanik, and M. Kristan, “A Low-Shot Object Counting Network With Iterative Prototype Adaptation,” in:Proceedings of the IEEE/CVF International Conference on Computer Vision , 2023, pp. 18872–18881

    2023

  47. [56]

    Faster R-CNN: Towards real- time object detection with region proposal networks,

    R. Girshick, S. Ren, K. He, and J. Sun, “Faster R-CNN: Towards real- time object detection with region proposal networks,”IEEE Transactions on Pattern Analysis and Machine Intelligence , vol. 39, no. 6, pp. 1137– 1149, 2016

    2016

  48. [57]

    Finding tiny faces,

    P. Hu and D. Ramanan, “Finding tiny faces,” in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2017, pp. 951–959

    2017

  49. [58]

    Recurrent attentive zooming for joint crowd counting and precise localization,

    C. Liu, X. Weng, and Y . Mu, “Recurrent attentive zooming for joint crowd counting and precise localization,” in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2019, pp. 1217–1226

    2019

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.