Pith. sign in

REVIEW 4 major objections 6 minor 42 references

Differential Alignment for Domain Adaptive Object Detection

T0 review · 4 major / 6 minor · reviewed 2026-08-11 · 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 a differential feature alignment strategy—weighting instance-level adversarial alignment by teacher-student prediction discrepancy and image-level alignment toward foreground regions—substantially improves domain…

desk verdict Solid incremental DAOD with strong reported numbers, but the headline ablation is confounded and the evaluation lacks variance reporting. read the letter →

arxiv 2412.12830 v1 pith:7YW4VQCJ submitted 2024-12-17 cs.CV

Xinyu He (1) , Xinhui Li (1) , Xiaojie Guo (1) ((1) College of Intelligence , Computing , Tianjin University , Tianjin , China) This is my paper · ORCID
classification cs.CV
keywords domainadaptiveobjectdetectiondifferentialfeaturealignmentpredictiondiscrepancyteacher-studentframeworkforeground-orientedadversariallearningpseudolabelsFasterR-CNN
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

Domain-adapted object detectors usually align source and target features with adversarial learning, giving every region the same alignment pressure. This paper tries to establish that a differential strategy works better: weigh each instance's alignment by how much the teacher and student disagree on it, and steer image-level alignment toward foreground regions while keeping background in the picture. It reports that this combination outperforms prior methods by 5.2 AP50 on Cityscapes to Foggy Cityscapes, 10.0 on Cityscapes to BDD100k daytime, and 4.2 on Sim10k to Cityscapes, with ablations showing each module contributes.

What carries the argument

Two weighting mechanisms on a teacher-student Faster R-CNN with adversarial discriminators. PDFA computes a prediction discrepancy matrix $P_{\mathrm{div}} = \mathrm{Square}(P_T - P_S)$ from teacher and student classification maps, averages over classes, min-max normalizes, and multiplies each instance's adversarial loss by its normalized weight; teacher proposal boxes are replaced by student ones so the two predictions can be compared. UFOA builds a foreground mask $M$ from ground-truth or pseudo-label boxes, splits the FPN P2 feature into $M \odot F_{\mathrm{img}}$ and $\bar{M} \odot F_{\mathrm{img}}$, feeds both to the image discriminator, and combines their losses as $\gamma \mathcal{L}_{\mathrm{adv}}^{\mathrm{fg}} + (1-\gamma) \mathcal{L}_{\mathrm{adv}}^{\mathrm{bg}}$. The full objective maximizes the two discriminators while minimizing the supervised and pseudo-label losses plus $\lambda$ times the two weighted adversarial losses.

What would settle it

Run the PDFA ablation with the prediction-discrepancy weights replaced by random or uniform weights while keeping every other component fixed; if AP50 on Cityscapes to Foggy Cityscapes stays near 57.3, the discrepancy signal itself is not doing the claimed work.

Watch

Extended reading notes

Core claim

The paper's central claim is that unequal attention—differential, not equal, feature alignment—improves domain adaptive object detection. It introduces PDFA, which converts teacher-student prediction discrepancies into per-instance weights for the adversarial alignment loss, and UFOA, which uses a pseudo-label-derived foreground mask to split the image feature into foreground and background and aligns them with a tunable balance. On the paper's own results, the full system reaches 57.3 AP50 on Cityscapes to Foggy Cityscapes versus 52.1 for the previous best REACT, 45.8 on Cityscapes to BDD100k daytime versus 35.8 for REACT, and 69.7 on Sim10k to Cityscapes, and the ablations attribute the gains to both modules.

Load-bearing premise

The load-bearing assumption is that the amount of disagreement between the teacher's and the student's predictions on an instance reliably indicates how much domain-specific information that instance carries, so that weighting alignment by this disagreement helps.

Editorial extensions

If this is right

  • If the central claim holds, equal-attention adversarial alignment is leaving measurable performance on the table: switching the instance module to discrepancy weighting adds 2.1 AP50, and the balanced foreground-oriented image alignment adds 1.1 AP50 on Cityscapes to Foggy Cityscapes.
  • The full method claims 57.3 AP50 on Cityscapes to Foggy Cityscapes, 45.8 on Cityscapes to BDD100k daytime, and 69.7 on Sim10k to Cityscapes, each exceeding the best previous result by at least 4.2 points.
  • The method should transfer across different sources of domain shift, since the same recipe works for weather, synthetic-to-real, and small-to-large dataset gaps.
  • Discarding background alignment entirely is harmful: the $\gamma=1.0$ foreground-only setting scores 55.6 AP50 versus 57.3 at $\gamma=0.8$, so a balanced foreground-background mix is part of the claimed gain.
  • The teacher-student discrepancy can be used as an automatic weighting signal even when no target-domain annotations are available, since it requires only the two model predictions.

Reading between the lines

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

  • Editorial inference: the prediction-discrepancy signal is a generalizable attention mechanism that could be lifted onto transformer detectors or dense prediction tasks, but the paper only demonstrates it on Faster R-CNN, as its own limitation section notes.
  • Editorial inference: because UFOA builds foreground masks from teacher pseudo-labels, missed objects in hard regions would be excluded from foreground priority, potentially reinforcing the blind spots that PDFA is meant to fix; a testable variant would use unsupervised saliency instead of pseudo-label boxes.
  • Editorial inference: the monotonic link between discrepancy magnitude and domain-shift severity is asserted with visual examples rather than measured; a direct per-instance measurement (e.g., pairing discrepancy with fog density) would tell whether PDFA is a true proxy or a useful heuristic.
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 differential alignment strategy for domain adaptive object detection (DAOD), built on an adaptive teacher-student framework with adversarial alignment. It introduces two modules: PDFA, which weights instance-level alignment according to teacher-student prediction discrepancy, and UFOA, which weights foreground versus background image-level alignment using an uncertainty factor γ. The method is evaluated on three benchmarks (Cityscapes→Foggy Cityscapes, Sim10k→Cityscapes, Cityscapes→BDD100k-daytime), reporting substantial improvements over prior work, including 57.3% AP50 on the first benchmark versus 52.1% for REACT. Ablation studies in Table 4 attribute gains to each module, and a supplementary section provides implementation details and qualitative visualizations.

Significance. If the reported results hold, the proposed differential alignment idea is a useful contribution to the DAOD literature: it moves beyond uniform alignment and directly addresses the uneven distribution of domain-specific information across instances and regions. The paper includes a GitHub link, a supplementary with implementation details for PDFA, and qualitative analyses. The improvements over strong baselines are large and category-wise consistent. However, the current experimental protocol has two load-bearing weaknesses: the central ablation is confounded by the unlisted burn-in/mutual-learning status, and the hyperparameter γ is tuned on the same benchmark that is later reported as the headline result. These issues prevent the paper from being accepted in its current form, but they are addressable with additional experiments and reporting.

major comments (4)
  1. [Ablation Studies / Table 4] Table 4, the central evidence for the contribution of PDFA and UFOA, is confounded by the burn-in phase and mutual-learning process. The baseline is defined as 'excluding the burn-in phase and mutual learning training process,' while the Implementation Details state that the final model is trained for 25k iterations including 10,000 burn-in iterations and 15,000 mutual-learning iterations. Table 4 has only columns for Strong Aug, PDFA, and UFOA, with no column indicating whether burn-in/mutual learning is enabled. If these components are present only in the final row, the 2.1% and 1.1% gains attributed to PDFA and UFOA are not isolated, and the full 57.3% versus 53.9% gap may be partly due to the training schedule. The authors must re-run the ablation with burn-in and mutual learning either enabled in every row or disabled in every row, or explicitly add columns for these components.
  2. [Ablation on uncertainty factor γ / Table 5] The hyperparameter γ is tuned on the same benchmark used for the headline result. Table 5 sweeps γ on Cityscapes→Foggy Cityscapes, and the value γ=0.8, which gives the best AP50 (57.3%), is then used when reporting the final method. Since the selection and reporting are both on the same target distribution, the 57.3% result is a selected, not a fixed-configuration, result. The supplementary acknowledges that γ=0.8 is not necessarily optimal but does not resolve the selection-on-test issue. Please select γ without access to target labels (e.g., on a source-only validation set) or present the full γ sweep in the main comparison so readers can see the sensitivity.
  3. [Experimental Validation / Tables 1–5] All reported numbers are single runs with no variance or significance statistics. Table 4 shows module-level differences of 1–2%, and Table 5 shows γ-dependent differences of 0.2–1.7%, which could be within run-to-run stochastic variation in training. The paper should report mean±std over at least three seeds for the main results and ablations, or otherwise justify why single runs are sufficient given the magnitude of the claimed differences.
  4. [Prediction-Discrepancy Feedback Alignment / Eqs. (2)–(6)] The core assumption that teacher-student prediction discrepancy is a faithful, monotonic proxy for domain-specific information is asserted with visual motivation (Fig. 3) but not quantitatively validated. Since this assumption drives the entire PDFA module, the paper would be substantially strengthened by a direct analysis, e.g., correlating per-instance discrepancy with a measurable domain-shift indicator (fog density, annotation noise, or category-level error rates) to show that higher discrepancy indeed corresponds to more domain-specific content.
minor comments (6)
  1. [Abstract] The phrase 'we investigates' should be corrected to 'we investigate'.
  2. [Table 2 caption] There is a typo in the caption: 'inbold' should be 'in bold', and a space is missing between 'in' and 'bold' throughout the caption.
  3. [Equations (5) and (8)] The notation for the domain flag is inconsistent: Eq. (5) uses d ∈ {0,1}^{N×1}, while Eq. (8) uses d ∈ {0,1}. Please clarify whether these are the same quantity and define the vector scalar distinction explicitly.
  4. [Supplementary Material, Implementation Details for PDFA] The supplementary states that teacher proposal boxes are replaced with student boxes when computing Eq. (2); the main text says the discrepancy is computed on 'the same instances.' Please reconcile this in the method section so the reader knows the actual implementation.
  5. [References] The reference list contains two entries labeled Zhu et al. 2020a and 2020b that appear to cite the same paper (Deformable DETR). If two different papers are intended, the citations should be distinguished; otherwise, please remove the duplicate.
  6. [Figure 2] The overview figure would be easier to follow if the caption defined the tensor dimensions N, C, and D, and if the symbols ⊙, ⊕, and ⊖ were explicitly described in the caption.

Circularity Check

2 steps flagged · score 4.0 of 10

Gamma is tuned on the Foggy Cityscapes benchmark and the ablation of PDFA/UFOA is not controlled for burn-in/mutual learning, so the headline Foggy result and module attributions are partly fitted; independent results on Sim10k and BDD keep the paper from being fully circular.

  1. fitted input called prediction [Table 5 and 'Ablation on uncertainty factor γ' (Eq. 9)]
    "The results of different γ in Eq. (9) is shown in Tab. 5, which highlights the effectiveness of our foreground-oriented alignment module. ... Table 5: Effect of the hyper-paramter γ in UFOA on adaptation from Cityscapes to Foggy Cityscapes. γ 0 0.5 0.8 1.0 APval50 55.1 55.9 57.3 55.6."

    The 57.3 AP50 headline on Cityscapes→Foggy Cityscapes (Table 1) is exactly the γ=0.8 row of Table 5, and Table 5 is itself an ablation on that same benchmark. The paper selects γ by comparing AP50 on Foggy Cityscapes and then reports the selected value as the method's best result; the 'prediction' is therefore the maximum of a one-parameter fit on the evaluation set, not an independent outcome. The supplementary's caveat that γ=0.8 is 'not necessarily the optimal value' does not remove the fact that the reported number was chosen from this table. Sim10k→Cityscapes and Cityscapes→BDD100k use the same fixed γ and are more independent, which limits the severity.

  2. other [Ablation Studies, Table 4; Implementation Details, Optimization]
    "the baseline consists of a teacher-student framework, image-level and instance-level alignment modules, without the differential attention mechanisms introduced by PDFA and UFOA, as well as excluding the burn-in phase and mutual learning training process. ... train the network for 25k iterations in total, including 10,000 iterations for burn-in and 15,000 iterations for teacher-student mutual learning."

    Table 4 is the paper's evidence that PDFA and UFOA each contribute (2.1% and 1.1% AP50). But the baseline row excludes burn-in and mutual learning, while the full training schedule in Implementation Details includes a 10k burn-in phase and a 15k mutual-learning phase. The table has no column for burn-in/mutual-learning status, so the rows marked +PDFA and +UFOA are not controlled against the baseline: any gain attributed to these modules may also be caused by enabling burn-in and mutual learning. Thus the claimed module contributions are not identified by the reported ablation; the full-vs-baseline gap likewise overstates the joint benefit of PDFA+UFOA if those training components are not held fixed.

full rationale

The paper is an empirical systems paper, not a derivation, and most of its pipeline (Eqs. 1–11) does not assume the target AP numbers. I find no self-citation chain, no imported uniqueness theorem, and no ansatz smuggled in via citation: the weight formulas in PDFA and the foreground mask in UFOA are defined from teacher–student outputs and pseudo-labels, not from the reported benchmarks. The two serious evidential weaknesses are (i) the uncertainty factor γ is selected on Cityscapes→Foggy Cityscapes, and the 57.3 AP50 headline in Table 1 is exactly the γ=0.8 row of that same benchmark's Table 5, making the headline result partially a one-parameter fit; and (ii) the Table 4 ablation baseline explicitly excludes burn-in and mutual learning, while the full schedule includes a 10k burn-in and a 15k mutual-learning phase, so the 2.1%/1.1% increments attributed to PDFA/UFOA are confounded by those unlisted components. The premise that prediction discrepancy tracks domain-specific information is an unproven modeling assumption but not a circularity, and the Sim10k and BDD100k results with fixed γ provide partial independent grounding. Overall circularity is therefore moderate: the central Foggy number and module attributions are partly fitted/confounded, but the method's general architecture is not defined in terms of its outputs.

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

No new physical or theoretical entities are introduced. The method rests on two empirical assumptions about image content (discrepancy as domain-richness signal, pseudo-label boxes as foreground proxy) plus the inherited assumption that adversarial alignment helps, and on three hand-set hyperparameters, one of which (gamma) is tuned on the target benchmark.

free parameters (3)
  • gamma (γ) = 0.8
    Uncertainty factor balancing foreground vs background image-level alignment loss (Eq. 9); tuned on Cityscapes->Foggy Cityscapes validation set.
  • lambda (λ) = 0.01
    Weight for the combined adversarial alignment losses in Eq. (10); chosen by hand, no sensitivity analysis reported.
  • alpha (α) = 0.9996
    EMA smoothing factor for teacher update in Eq. (1); set by hand.
assumptions (3)
  • domain assumption Teacher-student prediction discrepancy is a reliable indicator of the amount of domain-specific information in an instance region.
    Core premise behind PDFA weighting; motivated by examples in Fig. 3 but not proven.
  • domain assumption The union of teacher pseudo-label boxes on target data approximates the foreground region well enough for foreground/background feature separation.
    Used by UFOA mask generator; acknowledged to contain misclassification, mislocalization, and false detections.
  • domain assumption Adversarial feature alignment improves target-domain object detection.
    Inherited from the DAOD literature and used as the base learning paradigm.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Differential Alignment for Domain Adaptive Object Detection." pith.science (2026). https://pith.science/paper/7YW4VQCJ

@misc{pith2026241212830,
  author       = {Pith},
  title        = {Pith review of: Differential Alignment for Domain Adaptive Object Detection},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/7YW4VQCJ}},
  note         = {Machine review of arXiv:2412.12830}
}
read the original abstract

Domain adaptive object detection (DAOD) aims to generalize an object detector trained on labeled source-domain data to a target domain without annotations, the core principle of which is \emph{source-target feature alignment}. Typically, existing approaches employ adversarial learning to align the distributions of the source and target domains as a whole, barely considering the varying significance of distinct regions, say instances under different circumstances and foreground \emph{vs} background areas, during feature alignment. To overcome the shortcoming, we investigates a differential feature alignment strategy. Specifically, a prediction-discrepancy feedback instance alignment module (dubbed PDFA) is designed to adaptively assign higher weights to instances of higher teacher-student detection discrepancy, effectively handling heavier domain-specific information. Additionally, an uncertainty-based foreground-oriented image alignment module (UFOA) is proposed to explicitly guide the model to focus more on regions of interest. Extensive experiments on widely-used DAOD datasets together with ablation studies are conducted to demonstrate the efficacy of our proposed method and reveal its superiority over other SOTA alternatives. Our code is available at https://github.com/EstrellaXyu/Differential-Alignment-for-DAOD.

Figures

Figures reproduced from arXiv: 2412.12830 by the authors.

Figure 1
Figure 1. Different from previous methods adopting equal [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Overview of our method. Our approach is built upon the adaptive teacher-student framework. PDFA adjusts weights to different instances with respect to the discrepancy between predictions of the teacher and the student, while UFOA consists of a mask generator and an image-level discriminator. The mask generator produces a foreground-indicating mask to roughly separate the features of the last stage of the FPN into fo… view at source ↗
Figure 3
Figure 3. The proposals with top 2% prediction discrepan [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Visualization of pseudo labels generated by the [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 5
Figure 5. Figure 5: Feature distribution visualizations using PCA. Dif [PITH_FULL_IMAGE:figures/full_fig_p010_5.png]
Figure 6
Figure 6. Figure 6: Qualitative results on adaptation from Cityscapes to Foggy Cityscapes. Please zoom in for details. [PITH_FULL_IMAGE:figures/full_fig_p011_6.png]
Figure 7
Figure 7. Figure 7: Qualitative results on adaptation from Sim10k to Cityscapes. Please zoom in for details. [PITH_FULL_IMAGE:figures/full_fig_p011_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

42 extracted references · 26 canonical work pages

  1. [1]

    Abdi, H.; and Williams, L. J. 2010. Principal component analysis. Wiley interdisciplinary reviews: computational statistics, 2(4): 433--459

    work page 2010

  2. [2]

    Arpit, D.; Wang, H.; Zhou, Y.; and Xiong, C. 2022. Ensemble of averages: Improving model selection and boosting performance in domain generalization. Advances in Neural Information Processing Systems, 35: 8265--8277

    work page 2022

  3. [3]

    Cai, Q.; Pan, Y.; Ngo, C.-W.; Tian, X.; Duan, L.; and Yao, T. 2019. Exploring object relation in mean teacher for cross-domain detection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 11457--11466

    work page 2019

  4. [4]

    Carion, N.; Massa, F.; Synnaeve, G.; Usunier, N.; Kirillov, A.; and Zagoruyko, S. 2020. End-to-end object detection with transformers. In European conference on computer vision, 213--229. Springer

    2020

  5. [5]

    Chen, M.; Chen, W.; Yang, S.; Song, J.; Wang, X.; Zhang, L.; Yan, Y.; Qi, D.; Zhuang, Y.; Xie, D.; et al. 2022. Learning domain adaptive object detection with probabilistic teacher. arXiv preprint arXiv:2206.06293

    arXiv 2022

  6. [6]

    Chen, Y.; Li, W.; Sakaridis, C.; Dai, D.; and Van Gool, L. 2018. Domain adaptive faster r-cnn for object detection in the wild. In Proceedings of the IEEE conference on computer vision and pattern recognition, 3339--3348

    work page 2018

  7. [7]

    Chen, Y.; Wang, H.; Li, W.; Sakaridis, C.; Dai, D.; and Van Gool, L. 2021. Scale-aware domain adaptive faster r-cnn. International Journal of Computer Vision, 129(7): 2223--2243

    work page 2021

  8. [8]

    Cordts, M.; Omran, M.; Ramos, S.; Rehfeld, T.; Enzweiler, M.; Benenson, R.; Franke, U.; Roth, S.; and Schiele, B. 2016. The cityscapes dataset for semantic urban scene understanding. In Proceedings of the IEEE conference on computer vision and pattern recognition, 3213--3223

    2016

Show all 42 references
  1. [9]

    Deng, J.; Dong, W.; Socher, R.; Li, L.-J.; Li, K.; and Fei-Fei, L. 2009. Imagenet: A large-scale hierarchical image database. In 2009 IEEE conference on computer vision and pattern recognition, 248--255. Ieee

    2009

  2. [10]

    Deng, J.; Xu, D.; Li, W.; and Duan, L. 2023. Harmonious teacher for cross-domain object detection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 23829--23838

    2023

  3. [11]

    DeVries, T.; and Taylor, G. W. 2017. Improved regularization of convolutional neural networks with cutout. arXiv preprint arXiv:1708.04552

    2017 arXiv

  4. [12]

    Gao, C.; Liu, C.; Dun, Y.; and Qian, X. 2023. Csda: Learning category-scale joint feature for domain adaptive object detection. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 11421--11430

    2023

  5. [13]

    Girshick, R. 2015. Fast r-cnn. In Proceedings of the IEEE international conference on computer vision, 1440--1448

    2015

  6. [14]

    H.; and Chen, Q

    Gong, K.; Li, S.; Li, S.; Zhang, R.; Liu, C. H.; and Chen, Q. 2022. Improving transferability for domain adaptive detection transformers. In Proceedings of the 30th ACM International Conference on Multimedia, 1543--1551

    2022

  7. [15]

    He, K.; Zhang, X.; Ren, S.; and Sun, J. 2016. Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, 770--778

    2016

  8. [16]

    Hoyer, L.; Dai, D.; Wang, H.; and Van Gool, L. 2023. MIC: Masked image consistency for context-enhanced domain adaptation. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 11721--11732

    2023

  9. [17]

    Hsu, C.-C.; Tsai, Y.-H.; Lin, Y.-Y.; and Yang, M.-H. 2020. Every pixel matters: Center-aware feature alignment for domain adaptive object detector. In Computer Vision--ECCV 2020: 16th European Conference, Glasgow, UK, August 23--28, 2020, Proceedings, Part IX 16, 733--748. Springer

    2020

  10. [18]

    Huang, W.-J.; Lu, Y.-L.; Lin, S.-Y.; Xie, Y.; and Lin, Y.-Y. 2022. AQT: Adversarial Query Transformers for Domain Adaptive Object Detection. In IJCAI, 972--979

    2022

  11. [19]

    Jiang, J.; Chen, B.; Wang, J.; and Long, M. 2021. Decoupled adaptation for cross-domain object detection. arXiv preprint arXiv:2110.02578

    2021 arXiv

  12. [20]

    Johnson-Roberson, M.; Barto, C.; Mehta, R.; Sridhar, S.; Rosaen, K.; and Vasudevan, R. 2016. Driving in the Matrix: Can Virtual Worlds Replace Human-Generated Annotations for Real World Tasks? arXiv: Computer Vision and Pattern Recognition,arXiv: Computer Vision and Pattern Re...

    2016

  13. [21]

    Li, H.; Zhang, R.; Yao, H.; Zhang, X.; Hao, Y.; Song, X.; and Li, L. 2024. REACT: Remainder Adaptive Compensation for Domain Adaptive Object Detection. IEEE Transactions on Image Processing

    2024

  14. [22]

    Li, P.; Chen, X.; and Shen, S. 2019. Stereo r-cnn based 3d object detection for autonomous driving. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 7644--7652

    2019

  15. [23]

    Li, W.; Liu, X.; Yuan, Y.; and Bob. 2022 a . Sigma: Semantic-complete graph matching for domain adaptive object detection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 5291--5300

    2022

  16. [24]

    Li, Y.-J.; Dai, X.; Ma, C.-Y.; Liu, Y.-C.; Chen, K.; Wu, B.; He, Z.; Kitani, K.; and Vajda, P. 2022 b . Cross-domain adaptive teacher for object detection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 7581--7590

    2022

  17. [25]

    Lin, T.-Y.; Doll \'a r, P.; Girshick, R.; He, K.; Hariharan, B.; and Belongie, S. 2017. Feature pyramid networks for object detection. In Proceedings of the IEEE conference on computer vision and pattern recognition, 2117--2125

    2017

  18. [26]

    C.; and Marques, J

    Nascimento, J. C.; and Marques, J. S. 2006. Performance evaluation of object detection algorithms for video surveillance. IEEE Transactions on Multimedia, 8(4): 761--774

    2006

  19. [27]

    Redmon, J.; Divvala, S.; Girshick, R.; and Farhadi, A. 2016. You only look once: Unified, real-time object detection. In Proceedings of the IEEE conference on computer vision and pattern recognition, 779--788

    2016

  20. [28]

    Ren, S.; He, K.; Girshick, R.; and Sun, J. 2017. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. IEEE Transactions on Pattern Analysis & Machine Intelligence, 39(6): 1137--1149

    2017

  21. [29]

    Sakaridis, C.; Dai, D.; and Van Gool, L. 2018. Semantic foggy scene understanding with synthetic data. International Journal of Computer Vision, 126: 973--992

    2018

  22. [30]

    Tian, Z.; Shen, C.; Chen, H.; and He, T. 2020. FCOS: A simple and strong anchor-free object detector. IEEE transactions on pattern analysis and machine intelligence, 44(4): 1922--1933

    2020

  23. [31]

    A.; and Patel, V

    Vs, V.; Gupta, V.; Oza, P.; Sindagi, V. A.; and Patel, V. M. 2021. Mega-cda: Memory guided attention for category-aware unsupervised domain adaptive object detection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 4516--4526

    2021

  24. [32]

    Wang, W.; Cao, Y.; Zhang, J.; He, F.; Zha, Z.-J.; Wen, Y.; and Tao, D. 2021. Exploring sequence feature alignment for domain adaptive detection transformers. In Proceedings of the 29th ACM International Conference on Multimedia, 1730--1738

    2021

  25. [33]

    Xu, C.-D.; Zhao, X.-R.; Jin, X.; and Wei, X.-S. 2020. Exploring categorical regularization for domain adaptive object detection. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 11724--11733

    2020

  26. [34]

    Ye, M.; Shen, J.; Lin, G.; Xiang, T.; Shao, L.; and Hoi, S. C. 2021. Deep learning for person re-identification: A survey and outlook. IEEE transactions on pattern analysis and machine intelligence, 44(6): 2872--2893

    2021

  27. [35]

    Yu, F.; Chen, H.; Wang, X.; Xian, W.; Chen, Y.; Liu, F.; Madhavan, V.; and Darrell, T. 2020. BDD100K: A Diverse Driving Dataset for Heterogeneous Multitask Learning. In 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

    2020

  28. [36]

    Yu, J.; Liu, J.; Wei, X.; Zhou, H.; Nakata, Y.; Gudovskiy, D.; Okuno, T.; Li, J.; Keutzer, K.; and Zhang, S. 2022. MTTrans: Cross-domain object detection with mean teacher transformer. In European Conference on Computer Vision, 629--645. Springer

    2022

  29. [37]

    Zhang, L.; Zhou, W.; Fan, H.; Luo, T.; and Ling, H. 2024. Robust domain adaptive object detection with unified multi-granularity alignment. IEEE Transactions on Pattern Analysis and Machine Intelligence

    2024

  30. [38]

    Zhao, Y.; Lv, W.; Xu, S.; Wei, J.; Wang, G.; Dang, Q.; Liu, Y.; and Chen, J. 2024. Detrs beat yolos on real-time object detection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 16965--16974

    2024

  31. [39]

    Zhao, Z.; Wei, S.; Chen, Q.; Li, D.; Yang, Y.; Peng, Y.; and Liu, Y. 2023. Masked retraining teacher-student framework for domain adaptive object detection. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 19039--19049

    2023

  32. [41]

    Zhu, X.; Su, W.; Lu, L.; Li, B.; Wang, X.; and Dai, J. 2020 b . Deformable detr: Deformable transformers for end-to-end object detection. arXiv preprint arXiv:2010.04159

    2020 arXiv

  33. [42]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all...

  34. [43]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.