Pith. sign in

REVIEW 3 major objections 5 minor 54 references

Boosting Single-domain Generalized Object Detection via Vision-Language Knowledge Interaction

T0 review · 3 major / 5 minor · reviewed 2026-08-16 · deepseek-v4-flash

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

Pith's one-line read A detector trained on one domain can generalize to unseen corruptions by aligning each region with fine-grained text prompts from CLIP, gaining 8.8 mPC on Cityscapes-C and 7.9 mPC on DWD.

desk verdict Plausible region-level CLIP alignment for S-DGOD, but the central loss is underspecified as written and the numbers need code or errata before they can be trusted. read the letter →

arxiv 2504.19086 v1 pith:IJW5S37D submitted 2025-04-27 cs.CV

Xiaoran Xu , Jiangang Yang , Wenyue Chong , Wenhui Shi , Shichu Sun , Jing Xing , Jian Liu This is my paper · ORCID
classification cs.CV
keywords single-domaingeneralizedobjectdetectionvision-languagemodelscontrastivelearningdomain-invariantfeaturesregion-leveldataaugmentationrobustnesscross-modalalignment
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 is trying to show that a detector trained on a single source domain can generalize to unseen corruptions and weather by aligning each detected image region with a fine-grained text prompt from a frozen vision-language model. The authors claim their combination of a region-level contrastive loss (CRFI) and a proposal refining/mixing module (CPRM) produces domain-invariant region features, lifting mean performance under corruption (mPC) by 8.8 points on Cityscapes-C and 7.9 points on DWD over a plain Faster R-CNN baseline, and surpassing prior state-of-the-art methods. If right, this establishes region-level, cross-modal alignment as a stronger inductive bias than the scene-level prompts used by earlier VLM-based detectors.

What carries the argument

The central object is the Cross-modal and Region-aware Feature Interaction (CRFI) loss, an InfoNCE-based contrastive objective over object and background regions paired with class-aware text prompts. It is complemented by Cross-domain Proposal Refining and Mixing (CPRM), which averages RPN losses on clean and augmented images and concatenates their proposals. Together they make the text encoder's region-level semantics the training anchor for domain invariance.

What would settle it

Train the same pipeline replacing the CLIP text encoder with a fixed random projection of class labels (or with image-free one-hot vectors), keeping all other components. If the mPC gains on Cityscapes-C mostly survive, the claimed role of vision-language knowledge is wrong; if they collapse, the claim is supported. A second check: evaluate on a corruption type not represented in the color and frequency augmentations, such as severe JPEG compression or elastic distortion at high severity, and see whether the robustness margin over the baseline persists.

Watch

Extended reading notes

Core claim

The paper claims that simultaneously enforcing two kinds of invariance — between clean and augmented image regions, and between image regions and their textual descriptions — yields a detector that transfers to unseen domains. The CRFI loss uses InfoNCE to pull together region visual features and frozen CLIP text embeddings of prompts like 'a pedestrian on a sunny day' while separating different categories, and to align the same region before and after color and frequency augmentation. CPRM refines RPN proposals on both clean and augmented views and mixes them so the RPN learns consistent localization. The combined objective L = L_roi + L_CPRM + alpha * L_CRFI is what the experiments credit for the gains.

Load-bearing premise

The method assumes that frozen CLIP text embeddings for class labels and scenes remain a reliable, domain-invariant target for region features even when the visual appearance is heavily corrupted or augmented; if that text–image alignment breaks under the target shifts, the contrastive loss pulls features toward an unreliable anchor and the gains should shrink.

Editorial extensions

If this is right

  • Region-level text alignment can be plugged into any region-based detector, not just Faster R-CNN, without changing inference cost.
  • The gain pattern (largest on noise and weather) suggests the mechanism is most effective for appearance-level, global shifts.
  • CRFI and CPRM are complementary: the ablation shows CRFI supplies most of the gain and CPRM adds localization robustness.
  • The mPC metric, averaging mAP over corruptions and severities, is the evaluation protocol that determines the method's ranking.

Reading between the lines

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

  • If CLIP embeddings are the bottleneck, scaling the text encoder or using per-category attribute prompts could further improve generalization without new target data.
  • The proposal-mixing idea may transfer to semi-supervised or test-time adaptation settings where a teacher and student see different augmentations.
  • One could test CPRM alone on vision-only baselines such as PhysAug to isolate whether localization alignment helps when feature invariance is already strong.
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

3 major / 5 minor

Summary. The paper proposes a method for Single-Domain Generalized Object Detection (S-DGOD) that combines two components: Cross-modal and Region-aware Feature Interaction (CRFI) and Cross-domain Proposal Refining and Mixing (CPRM). CRFI uses a frozen CLIP text encoder to compute contrastive losses between detector region-of-interest features and text-prompt embeddings for both object and background regions, across clean and augmented images. CPRM refines and mixes region proposals from clean and augmented views to improve localization. The method is evaluated on Cityscapes-C and DWD, reporting state-of-the-art mPC improvements of +8.8% over the baseline and +1.6% over the previous best on Cityscapes-C, and +7.9% over baseline and +0.6% over previous best on DWD. Ablations on Cityscapes-C attribute large gains to CRFI (+8.5 mPC) and additional gains to CPRM.

Significance. If the claims hold, the paper makes a useful contribution by showing that region-level vision-language knowledge interaction can improve single-domain generalized object detection beyond scene-level prompt guidance. The core ideas—using frozen CLIP text embeddings as anchors for RoI features and adding proposal mixing across augmented domains—are well motivated and the ablations indicate substantial effect sizes. The manuscript also clearly identifies the limitation of prior VLM-based methods that only use coarse scene-level text. However, the significance is currently tempered by reproducibility gaps: no code is released, the augmentation details are in an unavailable supplementary, no error bars are reported, and the central CRFI loss is underspecified as written. The paper's central claim therefore rests on details that the reader cannot verify from the manuscript alone.

major comments (3)
  1. [Section 3.3, Eqs. (5)–(6)] The InfoNCE loss in Eqs. (5)–(6) is applied directly between region visual features I_obj, I_bg (extracted by the detector backbone f) and text features T_obj, T_bg (from the frozen CLIP text encoder g). The paper specifies no projection layer, no normalization, and no dimension matching between these two feature spaces. Faster R-CNN/FPN RoI features are not in CLIP's shared image–text embedding space, so the cosine-similarity or dot-product used by InfoNCE is not well defined as written. Because the ablation in Table 3 attributes +8.5 mPC to CRFI, this is a load-bearing technical detail: either a trainable projection is applied and omitted from the text, or the loss cannot be computed. Please specify the projection architecture, normalization, and how the InfoNCE similarity is computed, and ideally release code to confirm.
  2. [Table 2] There is an internal inconsistency in the DWD results. For the Baseline row, the four non-Daytime columns (Sunny 37.5, Night Sunny 29.2, Dusk Rainy 14.6, Night Rainy 33.1) have a mean of 28.6, not the reported mPC of 30.2. All other rows in Table 2 match the mean of their four non-Daytime columns. This suggests a transcription error in the baseline mPC. If the corrected baseline is 28.6, the reported +7.9 improvement over baseline changes. Please provide an erratum or corrected table, and ideally release exact per-corruption scores.
  3. [Section 4.1.3 / Section 4.2] The hyperparameter alpha is set to 0.01 on Cityscapes-C and 0.03 on DWD, but the paper does not state the validation procedure used to select these values, nor does it report sensitivity to alpha. Similarly, the random color and frequency augmentation parameters used in CRFI are said to be detailed 'in the supplementary material,' which is not available in the arXiv v1. Without this information, the reported improvements could reflect dataset-specific tuning. Please include the supplementary details, a sensitivity analysis for alpha, and ideally per-seed error bars or at least multiple-run standard deviations for the main tables.
minor comments (5)
  1. [Section 1, contributions list] The fourth bullet says CPRM 'signs and diversifies region proposals'; this appears to be a typo for 'aligns and diversifies.'
  2. [Algorithm 1 and Eq. (2)] The notation is inconsistent: Algorithm 1 returns P(C_obj) and P(C_bg), but Eq. (2) uses P_ori(C_obj), P_aug(C_obj), etc. Please define how the prompt template is split into original and augmented variants, and make the notation uniform (e.g., C_obj vs. C_O in the algorithm).
  3. [Section 3.4, Eq. (7)] It is unclear whether L_CPRM = L_refine exactly equals the standard RPN loss computed on clean and augmented images, or whether it is a modified loss. Please clarify the definition of L_aug_rpn and L_ori_rpn, and state how this replacement interacts with L_roi in Eq. (9).
  4. [Section 5.2 and Figure 6] The cosine similarity analysis is described only briefly. Please specify which features are compared, how the source and target features are pooled, and whether the plot shows per-corruption or averaged values.
  5. [General] The manuscript uses the 'Conference'17' LaTeX template, which leaves placeholder headers (e.g., 'Conference’17, July 2017') and causes some equations to be unreadably small. Please reformat to the target venue's style and ensure all equations are legible.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the central loss uses frozen CLIP text knowledge and external benchmarks; the only self-citation (PhysAug) is a baseline comparison, not a load-bearing premise.

full rationale

The derivation chain is self-contained. CRFI aligns detector ROI features to frozen CLIP text embeddings (Eqs. 2-6), where the text encoder is an external, fixed knowledge source; CPRM combines RPN losses and proposal concatenation (Eqs. 7-8) without using target-domain labels; the total loss in Eq. 9 is a training objective only. The SOTA claims rest on Cityscapes-C and DWD benchmark comparisons, which are external evaluations and are not used to fit the method's parameters. The only self-citation, PhysAug [42], appears as a comparative baseline and as a configuration reference, not as a load-bearing premise, uniqueness theorem, or forced-choice argument. No quoted equation reduces to its own input: Eq. 4 aligns augmented and clean visual features, Eqs. 5-6 align those with CLIP text features, and the reported alpha constants are not shown to be selected by target-domain performance. The noted dimensionality or projection issue in Eqs. 5-6 is a reproducibility/correctness concern, not a circularity one. Therefore no circular step is established.

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

The method rests on the assumption that frozen CLIP text embeddings provide a stable semantic anchor across visual domains, that color/frequency augmentation approximates the corruptions in Cityscapes-C and DWD, and that one selected region per category captures the class. These are domain assumptions, not derived from data in this paper.

free parameters (2)
  • alpha (loss weight for L_CRFI) = 0.01 (Cityscapes-C), 0.03 (DWD)
    Hand-set per benchmark; not derived from theory. Controls the balance between detection loss and cross-modal contrastive loss; the paper reports using different values for the two datasets without cross-validation details.
  • number of selected regions per image = 1 per unique category + 1 background
    Algorithm 1 selects only one object region per category and one background region per image. This is a design choice that affects the contrastive learning signal; the paper does not justify or ablate this number.
assumptions (3)
  • domain assumption Frozen CLIP text embeddings of class-label prompts provide a domain-invariant target for region-level visual features.
    Eq. 2 and the losses in Eq. 3-6 assume that text embeddings of prompts such as 'a pedestrian on a sunny day' are stable across domains and that aligning image region features to them improves generalization to unseen corruptions.
  • domain assumption Random color and frequency transformations create pseudo-target images that are representative of the corruptions in Cityscapes-C and DWD.
    Section 4.1.3 states the augmentation used for CRFI and CPRM. The cross-domain interaction relies on these augmented images as surrogate target domains; if the augmentation distribution is far from the actual corruptions, the learned invariance may not transfer.
  • domain assumption The single selected region per category is a representative positive sample for that category.
    Algorithm 1 picks one object region per unique category per image. The contrastive loss treats that region as the positive anchor, which assumes the region is semantically unambiguous and not an outlier.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Boosting Single-domain Generalized Object Detection via Vision-Language Knowledge Interaction." pith.science (2026). https://pith.science/paper/IJW5S37D

@misc{pith2026250419086,
  author       = {Pith},
  title        = {Pith review of: Boosting Single-domain Generalized Object Detection via Vision-Language Knowledge Interaction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IJW5S37D}},
  note         = {Machine review of arXiv:2504.19086}
}
read the original abstract

Single-Domain Generalized Object Detection~(S-DGOD) aims to train an object detector on a single source domain while generalizing well to diverse unseen target domains, making it suitable for multimedia applications that involve various domain shifts, such as intelligent video surveillance and VR/AR technologies. With the success of large-scale Vision-Language Models, recent S-DGOD approaches exploit pre-trained vision-language knowledge to guide invariant feature learning across visual domains. However, the utilized knowledge remains at a coarse-grained level~(e.g., the textual description of adverse weather paired with the image) and serves as an implicit regularization for guidance, struggling to learn accurate region- and object-level features in varying domains. In this work, we propose a new cross-modal feature learning method, which can capture generalized and discriminative regional features for S-DGOD tasks. The core of our method is the mechanism of Cross-modal and Region-aware Feature Interaction, which simultaneously learns both inter-modal and intra-modal regional invariance through dynamic interactions between fine-grained textual and visual features. Moreover, we design a simple but effective strategy called Cross-domain Proposal Refining and Mixing, which aligns the position of region proposals across multiple domains and diversifies them, enhancing the localization ability of detectors in unseen scenarios. Our method achieves new state-of-the-art results on S-DGOD benchmark datasets, with improvements of +8.8\%~mPC on Cityscapes-C and +7.9\%~mPC on DWD over baselines, demonstrating its efficacy.

Figures

Figures reproduced from arXiv: 2504.19086 by the authors.

Figure 1
Figure 1. This figure provides a high-level comparison of [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. This figure illustrates our CRFI, which extracts multi-region invariant features through text-image interaction. [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. This figure illustrates our CPRM, which enables [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: The visualizations of the baseline, PhysAug, and our model on the Cityscapes-C dataset are presented. These include [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]
Figure 5
Figure 5. Figure 5: Visualization of Daytime Foggy in DWD. of the DWD test set, as shown in [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 6
Figure 6. Figure 6: The plots show cosine similarity between the Cityscapes validation set and Cityscapes-C at severity 1 and severity 5, [PITH_FULL_IMAGE:figures/full_fig_p009_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

54 extracted references · 36 canonical work pages

  1. [1]

    Sungha Choi, Sanghun Jung, Huiwon Yun, Joanne T Kim, Seungryong Kim, and Jaegul Choo. 2021. Robustnet: Improving domain generalization in urban-scene segmentation via instance selective whitening. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 11580–11590

    work page 2021

  2. [2]

    Ekin D Cubuk, Barret Zoph, Jonathon Shlens, and Quoc V Le. 2020. Randaug- ment: Practical automated data augmentation with a reduced search space. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition workshops. 702–703

    work page 2020

  3. [3]

    Muhammad Sohail Danish, Muhammad Haris Khan, Muhammad Akhtar Munir, M Saquib Sarfraz, and Mohsen Ali. 2024. Improving Single Domain-Generalized Object Detection: A Focus on Diversification and Alignment. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 17732–17742

    work page 2024

  4. [4]

    Karan Desai and Justin Johnson. 2021. Virtex: Learning visual representations from textual annotations. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 11162–11173

    work page 2021

  5. [5]

    Terrance DeVries and Graham W Taylor. 2017. Improved regularization of convolutional neural networks with cutout. arXiv preprint arXiv:1708.04552 (2017)

    arXiv 2017

  6. [6]

    Jeffrey Donahue, Lisa Anne Hendricks, Sergio Guadarrama, Marcus Rohrbach, Subhashini Venugopalan, Kate Saenko, and Trevor Darrell. 2015. Long-term recurrent convolutional networks for visual recognition and description. In Proceedings of the IEEE conference on computer vision and pattern recognition . 2625–2634

    work page 2015

  7. [7]

    Mohammad Fahes, Tuan-Hung Vu, Andrei Bursuc, Patrick Pérez, and Raoul De Charette. 2023. Poda: Prompt-driven zero-shot domain adaptation. InProceed- ings of the IEEE/CVF International Conference on Computer Vision. 18623–18633. Boosting Single-Domain Generalized Object Detection via Vision-Language Knowledge Interaction Conference’17, July 2017, Washingto...

    work page 2023

  8. [8]

    Christoph Feichtenhofer, Axel Pinz, and Andrew Zisserman. 2017. Detect to track and track to detect. In Proceedings of the IEEE international conference on computer vision. 3038–3046

    work page 2017

Show all 54 references
  1. [9]

    Robert Geirhos, Patricia Rubisch, Claudio Michaelis, Matthias Bethge, Felix A Wichmann, and Wieland Brendel. 2018. ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness. arXiv preprint arXiv:1811.12231 (2018)

    2018 arXiv

  2. [10]

    Mahmoud Hassaballah, Mourad A Kenk, Khan Muhammad, and Shervin Minaee

  3. [11]

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. 2016. Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition. 770–778

    2016

  4. [12]

    Dan Hendrycks and Thomas Dietterich. 2019. Benchmarking neural net- work robustness to common corruptions and perturbations. arXiv preprint arXiv:1903.12261 (2019)

    2019 arXiv

  5. [13]

    Dan Hendrycks and Thomas G Dietterich. 2018. Benchmarking neural net- work robustness to common corruptions and surface variations. arXiv preprint arXiv:1807.01697 (2018)

    2018 arXiv

  6. [14]

    Dan Hendrycks, Norman Mu, Ekin D Cubuk, Barret Zoph, Justin Gilmer, and Balaji Lakshminarayanan. 2019. Augmix: A simple data processing method to improve robustness and uncertainty. arXiv preprint arXiv:1912.02781 (2019)

    2019 arXiv

  7. [15]

    Lei Huang, Yi Zhou, Fan Zhu, Li Liu, and Ling Shao. 2019. Iterative normaliza- tion: Beyond standardization towards efficient whitening. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 4874–4883

    2019

  8. [16]

    Christoph Hümmer, Manuel Schwonberg, Liangwei Zhou, Hu Cao, Alois Knoll, and Hanno Gottschalk. 2024. Strong but Simple: A Baseline for Domain General- ized Dense Perception by CLIP-Based Transfer Learning. In Proceedings of the Asian Conference on Computer Vision. 4223–4244

    2024

  9. [17]

    Andrej Karpathy and Li Fei-Fei. 2015. Deep visual-semantic alignments for generating image descriptions. In Proceedings of the IEEE conference on computer vision and pattern recognition. 3128–3137

    2015

  10. [18]

    Prannay Khosla, Piotr Teterwak, Chen Wang, Aaron Sarna, Yonglong Tian, Phillip Isola, Aaron Maschinot, Ce Liu, and Dilip Krishnan. 2020. Supervised contrastive learning. Advances in neural information processing systems 33 (2020), 18661– 18673

    2020

  11. [19]

    Wooju Lee, Dasol Hong, Hyungtae Lim, and Hyun Myung. 2024. Object-Aware Domain Generalization for Object Detection. In Proceedings of the AAAI Confer- ence on Artificial Intelligence, Vol. 38. 2947–2955

    2024

  12. [20]

    Deng Li, Aming Wu, Yaowei Wang, and Yahong Han. 2024. Prompt-driven dynamic object-centric learning for single domain generalization. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 17606– 17615

    2024

  13. [21]

    Tsung-Yi Lin, Piotr Dollár, Ross Girshick, Kaiming He, Bharath Hariharan, and Serge Belongie. 2017. Feature pyramid networks for object detection. In Proceed- ings of the IEEE conference on computer vision and pattern recognition. 2117–2125

    2017

  14. [22]

    Yajing Liu, Shijun Zhou, Xiyao Liu, Chunhui Hao, Baojie Fan, and Jiandong Tian

  15. [23]

    Claudio Michaelis, Benjamin Mitzkus, Robert Geirhos, Evgenia Rusak, Oliver Bringmann, Alexander S Ecker, Matthias Bethge, and Wieland Brendel. 2019. Benchmarking robustness in object detection: Autonomous driving when winter is coming. arXiv preprint arXiv:1907.07484 (2019)

    2019 arXiv

  16. [24]

    Apostolos Modas, Rahul Rade, Guillermo Ortiz-Jiménez, Seyed-Mohsen Moosavi- Dezfooli, and Pascal Frossard. 2022. Prime: A few primitives can boost robustness to common corruptions. In European Conference on Computer Vision. Springer, 623–640

    2022

  17. [25]

    Richard A Newcombe, Shahram Izadi, Otmar Hilliges, David Molyneaux, David Kim, Andrew J Davison, Pushmeet Kohi, Jamie Shotton, Steve Hodges, and Andrew Fitzgibbon. 2011. Kinectfusion: Real-time dense surface mapping and tracking. In 2011 10th IEEE international symposium on mi...

    2011

  18. [26]

    Xingang Pan, Xiaohang Zhan, Jianping Shi, Xiaoou Tang, and Ping Luo. 2019. Switchable whitening for deep representation learning. In Proceedings of the IEEE/CVF international conference on computer vision. 1863–1871

    2019

  19. [27]

    Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, et al. 2021. Learning transferable visual models from natural language supervision. In International conference on machine learni...

    2021

  20. [28]

    Yongming Rao, Wenliang Zhao, Guangyi Chen, Yansong Tang, Zheng Zhu, Guan Huang, Jie Zhou, and Jiwen Lu. 2022. Denseclip: Language-guided dense predic- tion with context-aware prompting. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 18082–18091

    2022

  21. [29]

    Sylvestre-Alvise Rebuffi, Sven Gowal, Dan Andrei Calian, Florian Stimberg, Olivia Wiles, and Timothy A Mann. 2021. Data augmentation can improve robustness. Advances in Neural Information Processing Systems 34 (2021), 29935–29948

    2021

  22. [30]

    Joseph Redmon and Ali Farhadi. 2018. Yolov3: An incremental improvement. arXiv preprint arXiv:1804.02767 (2018)

    2018 arXiv

  23. [31]

    Shaoqing Ren, Kaiming He, Ross Girshick, and Jian Sun. 2015. Faster r-cnn: Towards real-time object detection with region proposal networks. Advances in neural information processing systems 28 (2015)

    2015

  24. [32]

    Christos Sakaridis, Dengxin Dai, and Luc Van Gool. 2018. Semantic foggy scene understanding with synthetic data. International Journal of Computer Vision 126 (2018), 973–992

    2018

  25. [33]

    Shuai Shen, Wanhua Li, Xiaobing Wang, Dafeng Zhang, Zhezhu Jin, Jie Zhou, and Jiwen Lu. 2023. Clip-cluster: Clip-guided attribute hallucination for face clustering. In Proceedings of the IEEE/CVF international conference on computer vision. 20786–20795

    2023

  26. [34]

    Connor Shorten and Taghi M Khoshgoftaar. 2019. A survey on image data augmentation for deep learning. Journal of big data 6, 1 (2019), 1–48

    2019

  27. [35]

    Bo Sun, Banghuai Li, Shengcai Cai, Ye Yuan, and Chi Zhang. 2021. Fsce: Few-shot object detection via contrastive proposal encoding. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition. 7352–7362. Conference’17, July 2017, Washington, DC, USA Xu et al

    2021

  28. [36]

    Puru Vaish, Shunxin Wang, and Nicola Strisciuglio. 2024. Fourier-basis func- tions to bridge augmentation gap: Rethinking frequency augmentation in image classification. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 17763–17772

    2024

  29. [37]

    Vidit Vidit, Martin Engilberge, and Mathieu Salzmann. 2023. Clip the gap: A single domain generalization approach for object detection. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 3219–3229

    2023

  30. [38]

    Xin Wang, Thomas E Huang, Benlin Liu, Fisher Yu, Xiaolong Wang, Joseph E Gonzalez, and Trevor Darrell. 2021. Robust object detection via instance-level temporal cycle confusion. In Proceedings of the IEEE/CVF International Conference on Computer Vision. 9143–9152

    2021

  31. [39]

    Thomas Whelan, Renato F Salas-Moreno, Ben Glocker, Andrew J Davison, and Stefan Leutenegger. 2016. ElasticFusion: Real-time dense SLAM and light source estimation. The International Journal of Robotics Research35, 14 (2016), 1697–1716

    2016

  32. [40]

    Aming Wu and Cheng Deng. 2022. Single-domain generalized object detection in urban scene via cyclic-disentangled self-distillation. In Proceedings of the IEEE/CVF Conference on computer vision and pattern recognition. 847–856

    2022

  33. [41]

    Fan Wu, Jinling Gao, Lanqing Hong, Xinbing Wang, Chenghu Zhou, and Nanyang Ye. 2024. G-NAS: Generalizable Neural Architecture Search for Single Domain Generalization Object Detection. In Proceedings of the AAAI Conference on Artifi- cial Intelligence, Vol. 38. 5958–5966

    2024

  34. [42]

    Xiaoran Xu, Jiangang Yang, Wenhui Shi, Siyuan Ding, Luqing Luo, and Jian Liu

  35. [43]

    Lewei Yao, Jianhua Han, Youpeng Wen, Xiaodan Liang, Dan Xu, Wei Zhang, Zhenguo Li, Chunjing Xu, and Hang Xu. 2022. Detclip: Dictionary-enriched visual-concept paralleled pre-training for open-world detection. Advances in Neural Information Processing Systems 35 (2022), 9125–9138

    2022

  36. [44]

    Fisher Yu, Haofeng Chen, Xin Wang, Wenqi Xian, Yingying Chen, Fangchen Liu, Vashisht Madhavan, and Trevor Darrell. 2020. Bdd100k: A diverse driving dataset for heterogeneous multitask learning. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognitio...

    2020

  37. [45]

    arXiv preprint arXiv:2412.11807 (2024)

    PhysAug: A Physical-guided and Frequency-based Data Augmentation for Single-Domain Generalized Object Detection. arXiv preprint arXiv:2412.11807 (2024)

    2024 arXiv

  38. [46]

    Long Zhao, Ting Liu, Xi Peng, and Dimitris Metaxas. 2020. Maximum-entropy adversarial data augmentation for improved generalization and robustness. Ad- vances in Neural Information Processing Systems 33 (2020), 14435–14447

    2020

  39. [47]

    Yuyang Zhao, Zhun Zhong, Na Zhao, Nicu Sebe, and Gim Hee Lee. 2022. Style- hallucinated dual consistency learning for domain generalized semantic segmen- tation. In European conference on computer vision. Springer, 535–552

    2022

  40. [48]

    Hongyi Zhang, Moustapha Cisse, Yann N Dauphin, and David Lopez-Paz. 2017. mixup: Beyond empirical risk minimization. arXiv preprint arXiv:1710.09412 (2017)

    2017 arXiv

  41. [49]

    Kaiyang Zhou, Jingkang Yang, Chen Change Loy, and Ziwei Liu. 2022. Conditional prompt learning for vision-language models. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 16816–16825

    2022

  42. [50]

    Kaiyang Zhou, Jingkang Yang, Chen Change Loy, and Ziwei Liu. 2022. Learning to prompt for vision-language models. International Journal of Computer Vision 130, 9 (2022), 2337–2348

    2022

  43. [51]

    Kaiyang Zhou, Ziwei Liu, Yu Qiao, Tao Xiang, and Chen Change Loy. 2022. Domain generalization: A survey. IEEE Transactions on Pattern Analysis and Machine Intelligence 45, 4 (2022), 4396–4415

    2022

  44. [54]

    Barret Zoph, Ekin D Cubuk, Golnaz Ghiasi, Tsung-Yi Lin, Jonathon Shlens, and Quoc V Le. 2020. Learning data augmentation strategies for object detection. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XXVII 16. Spring...

    2020

  45. [2020]

    IEEE transactions on intelligent transportation systems 22, 7 (2020), 4230–4242

    Vehicle detection and tracking in adverse weather using a deep learning framework. IEEE transactions on intelligent transportation systems 22, 7 (2020), 4230–4242

    2020

  46. [2024]

    In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Unbiased Faster R-CNN for Single-source Domain Generalized Object Detection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 28838–28847

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.