Pith. sign in

REVIEW 3 major objections 6 minor 38 references

AetherVision-Bench: An Open-Vocabulary RGB-Infrared Benchmark for Multi-Angle Segmentation across Aerial and Ground Perspectives

T0 review · 3 major / 6 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read AetherVision-Bench, a multi-angle RGB-infrared benchmark, shows that open-vocabulary segmentation models lose nearly half their mIoU when moving from RGB to infrared and generalize poorly across viewing angles.

desk verdict A useful benchmark idea with a real arithmetic flaw in its central claim: the 'nearly half' sensor-shift drop is inflated by comparing 6-dataset RGB averages to 4-dataset IR averages. read the letter →

arxiv 2506.03709 v1 pith:M2326VRZ submitted 2025-06-04 cs.CV

Aniruddh Sikdar , Aditya Gandhamal , Suresh Sundaram This is my paper
classification cs.CV
keywords open-vocabularysemanticsegmentationRGB-infraredbenchmarkmulti-anglesensormodalityshiftviewpointgeneralizationzero-shottransferaerialandgroundperspectivesembodiedAIperception
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

Open-vocabulary semantic segmentation (OVSS) assigns pixel labels from arbitrary text categories, but the paper asks whether such models survive the shifts that real drones and vehicles face: changing sensor modality from visible RGB to infrared, and changing viewing angle from bird's-eye to slant-angle to ground-level. To answer this, it introduces AetherVision-Bench, a collection of six RGB and four infrared datasets spanning those viewpoints, and evaluates four state-of-the-art OVSS models under three training regimes. The central finding is that sensor shift is a severe failure mode: mean intersection-over-union drops by nearly half when models move from RGB to IR, even on co-registered ground-level pairs. Models trained on one viewpoint also generalize poorly to the others, suggesting that downstream fine-tuning cannot replace diverse pretraining. The benchmark's contribution is a standardized protocol for measuring these two robustness axes, which the paper argues are essential for embodied-AI deployment.

What carries the argument

The central object is AetherVision-Bench itself, a benchmark organized by a three-way taxonomy of viewing angles — bird's-eye view, slant-angle, and ground-level — crossed with RGB versus infrared modality. Its machinery is a fixed evaluation protocol: train each model on one of three source settings (COCO-Stuff, ground-level MSRS RGB, or slant-angle IndraEye RGB), then measure mIoU on all six RGB and four IR datasets in the benchmark. The cross of training viewpoint with evaluation viewpoint and the paired RGB-IR datasets make sensor and viewpoint shifts measurable; the mIoU drop between a dataset's RGB and IR columns is the operational definition of sensor-shift fragility.

What would settle it

Take a single co-registered RGB-IR dataset with identical scenes and viewpoints, such as MSRS, and evaluate the same models after controlling class sets and resolution; if the RGB-to-IR mIoU drop disappears or shrinks sharply, the benchmark's attribution of the drop to sensor modality would be largely an artifact of dataset confounds rather than a general sensor-shift fragility.

Watch

Extended reading notes

Core claim

The paper's claim is that current open-vocabulary segmentation models are not robust to sensor modality or viewing-angle shifts, and that this brittleness is large enough to matter in practice. Across all three training settings, every evaluated model's mIoU falls by roughly half on infrared inputs relative to RGB, and the drop is worst on noisy multispectral video. Training on ground-level RGB does not help at aerial viewpoints, and training on slant-angle RGB does not help at ground level; models appear to lean on their pretraining to cover unseen viewpoints. The authors consequently conclude that sensor modality, viewing angle, and class semantics are the factors that dominate zero-shot transfer performance, and that these factors need to be measured explicitly by a dedicated benchmark.

Load-bearing premise

The benchmark's diagnosis assumes the six datasets differ mainly in viewing angle and sensor modality, but they also differ in scene content, class sets, resolution, annotation style, and sensor hardware, so part of the measured mIoU drop could come from those other differences.

Editorial extensions

If this is right

  • Deployed systems that need night or weather vision cannot currently rely on open-vocabulary segmentation over IR, since the near-halving of mIoU would produce unusable maps in low-light conditions.
  • Pretraining diversity matters more than fine-tuning data, because training on a single viewpoint leaves models unable to transfer to other viewpoints, so benchmark rankings should weight cross-viewpoint generalization rather than in-domain mIoU.
  • Co-registered RGB-IR pairs with the same scene and viewpoint still show the drop, meaning the loss is at least partly a representation gap in how models encode thermal input, not just a scene change.
  • The benchmark's three training settings give a standard protocol for testing fixes such as domain adaptation, RGB-to-IR translation, fusion, or viewpoint-conditioned prompts, with a clear before-and-after metric.
  • Performance on noisy multispectral video deteriorates even further, indicating that sensor noise compounds the modality shift.

Reading between the lines

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

  • The paper's datasets differ in scene content, class sets, resolution, and annotation style, so the measured sensor-shift and viewpoint-shift effects are likely partially confounded; a controlled sub-benchmark with matched classes and resolutions would let the field attribute the drop more precisely.
  • If the drop on co-registered RGB-IR pairs is a representation gap, then inserting an RGB-to-IR translation module before the segmenter and measuring how much mIoU it recovers would directly test that gap.
  • Viewpoint-specific text prompts, for example adding 'aerial view' or 'ground view' to class names, might recover part of the cross-view drop, since the paper finds textual similarity shapes performance; the benchmark could test this without retraining.
  • The near-halving result sets an upper bound on what current vision-language encoders contribute to infrared understanding, suggesting future progress may require pretraining on paired thermal-visible data rather than better segmentation decoders alone.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 6 minor

Summary. The paper introduces AetherVision-Bench, a benchmark composed of six public semantic segmentation datasets spanning bird's-eye, slant-angle, and ground-level viewpoints, with four of the datasets also providing infrared (IR) imagery. The authors evaluate four open-vocabulary semantic segmentation models (SAN, ZegFormer, OVSeg, CAT-Seg) under three training settings (COCO-Stuff, MSRS RGB, and IndraEye RGB) and report per-dataset mIoU tables. The central findings are that current OVSS models generalize poorly from RGB to IR and from one viewpoint to another, with the paper claiming performance decreases by nearly half on sensor shift. The benchmark addresses a relevant gap in the evaluation of OVSS models for embodied AI applications.

Significance. If the benchmark and its conclusions are properly supported, this work would be a useful testbed for the robustness of open-vocabulary segmentation models in aerial and ground robotics, a largely underexplored direction. The paper brings together six datasets with different viewpoints and modalities and evaluates four representative model architectures under multiple training settings, which is a genuine contribution. However, the reproducibility and quantitative rigor of the benchmark are currently limited by the missing evaluation protocol and by an inconsistent computation of the headline sensor-shift claim. The direction of the findings is credible, but the specific magnitude stated in Section 4.1 is not supported as reported.

major comments (3)
  1. [Section 4.1, Sensor Domain Shift] The evaluation protocol is not specified. Section 4 states only that models are evaluated using their official implementations and pretrained weights, with mIoU as the metric. The paper does not provide the class text prompts or prompt templates used for open-vocabulary evaluation, the input image resolution, the class-name mapping for each dataset, or the training hyperparameters for the three training settings (COCO-Stuff, MSRS, IndraEye). Because OVSS results are highly sensitive to prompt phrasing and resolution, the benchmark as presented is not reproducible, and the cross-dataset comparisons in Tables 2-4 mix unknown configurations. The full protocol (exact prompt strings, image sizes, number of iterations, learning rates, batch sizes, and any post-processing) must be included in the paper or a supplement.
  2. [Section 3.1, Dataset selection] The attribution of mIoU drops to sensor modality and viewing angle assumes the six datasets are comparable except along these two axes. In reality, ISPRS (aerial German urban scenes), UA Vid (drone over varied terrain), IndraEye (drone and ground), CART (Caltech aerial thermal), MSRS (ground-level fusion), and MVSeg (multispectral video) differ in scene content, class sets, annotation style, resolution, and sensor hardware. These confounds enter at dataset selection in Section 3.1 and affect all Section 4.1 conclusions. The paper should report the number of classes, class lists, and class overlap across datasets, and discuss how these differences are controlled or how their influence is bounded.
  3. [Tables 2-4 and Section 4] No variance or repeated-run statistics are reported. Each table shows a single mIoU value per model-dataset combination, so differences between models (e.g., Table 2: SAN RGB 27.79 vs. OVSeg 21.75) cannot be distinguished from training or evaluation noise. Given that the paper's purpose is to compare robustness across conditions, at least standard deviations over multiple seeds or bootstrap confidence intervals should be provided.
minor comments (6)
  1. [Title and Abstract] There is a typo in the title: 'Open-V ocabulary' should be 'Open-Vocabulary'.
  2. [Section 4.1, Sensor Domain Shift] The phrase 'even when models are trained on the MSRS RGB dataset, their performance on co-registered IR pairs drops to less than half' is supported only for the MSRS IR column in Table 3; the statement should be clarified to specify that this refers to the MSRS subset, not the overall average.
  3. [Section 1, Introduction] The word 'visiable' appears in 'both visiable (RGB) and infrared (IR) sensor data'; it should be 'visible'.
  4. [Section 4.1, Evaluation under diverse viewing angles] 'incase' should be 'in case'.
  5. [Figure 1] The radar chart in Figure 1 is described as illustrating the 'balanced nature' of the benchmark 'across these key dimensions,' but the axes (number of classes, resolution, total pixel volume, overall volume) are not defined in the text, and it is unclear how 'overall dataset volume' is normalized. A caption with axis definitions would improve clarity.
  6. [Section 5, Conclusion] The conclusion restates the contributions but does not mention any limitations of the benchmark, such as the lack of class-overlap control or the small number of IR datasets; adding a limitations paragraph would strengthen the paper.

Circularity Check

0 steps flagged · score 2.0 of 10

No load-bearing circularity: the benchmark findings are direct measurements, with only minor non-load-bearing self-citations.

full rationale

The paper contains no fitted parameters, no equations, and no first-principles derivation; the central sensor-shift and viewpoint-generalization claims are summaries of mIoU values obtained from official implementations of SAN, Zegformer, OVSeg, and CAT-Seg on the selected datasets. These measurements are externally grounded in the datasets and model checkpoints, and the conclusion that performance drops under RGB-to-IR shifts does not depend on a self-citation: Tables 2-4 provide the raw numbers from which the claim is read. The authors' self-citations ([1], [20]-[25], including the IndraEye dataset [24]) provide related-work context and one benchmark constituent, but none is invoked as proof of the robustness finding. The Section 4.1 statement that performance 'decreases by nearly half' is arithmetically fragile because the RGB mIoU column averages six datasets while the IR column averages four, a comparability issue that belongs to correctness risk rather than circularity; matched-pair drops remain substantial. Accordingly, no circular step satisfying the Eq.-X-equals-Eq.-Y criterion is identified.

Assumptions & free parameters 1 free parameters · 3 assumptions · 0 invented entities

The benchmark rests on three unproven domain assumptions: representativeness of the selected datasets, causal attribution of performance differences to the two axes of interest, and comparability of the evaluation across models. The free-parameter list is minimal; the biggest unmeasured variable is the prompt wording for open-vocabulary classes. No invented entities are needed.

free parameters (1)
  • Class text prompts used for open-vocabulary evaluation = Not specified
    Zero-shot OVSS mIoU is highly sensitive to the wording of class labels (e.g., 'building' versus 'house'); the paper does not report the prompts used for each dataset, so the absolute mIoU values and cross-dataset comparisons are not reproducible.
assumptions (3)
  • domain assumption The six selected datasets (ISPRS, UA Vid, IndraEye, CART, MSRS, MVSeg) are representative of aerial, slant-angle, and ground-level OVSS deployment conditions.
    Section 3.1 filters datasets based on viewing angle and sensor modality, but provides no evidence that these six datasets cover the range of real-world conditions (illumination, weather, camera hardware) faced by drones and ground vehicles.
  • domain assumption Differences in mIoU across datasets can be attributed primarily to sensor modality and viewing angle rather than to dataset-specific content or class distributions.
    The Section 4.1 discussion (Sensor Domain Shift, Evaluation under diverse viewing angles) interprets drop patterns as sensor and viewpoint effects, but the datasets are not matched on scene type, resolution, or class ontology, so the attribution is confounded.
  • domain assumption The evaluation protocol using official implementations, pretrained weights, and unspecified fine-tuning settings is comparable across all models and datasets.
    Section 4 states official implementations are used, but training on COCO-Stuff, MSRS, and IndraEye is not specified (prompts, iterations, learning rate), so the fairness of comparisons across models and settings is assumed.

how reviews work

0 comments
Cite this review

Pith. "Pith review of AetherVision-Bench: An Open-Vocabulary RGB-Infrared Benchmark for Multi-Angle Segmentation across Aerial and Ground Perspectives." pith.science (2026). https://pith.science/paper/M2326VRZ

@misc{pith2026250603709,
  author       = {Pith},
  title        = {Pith review of: AetherVision-Bench: An Open-Vocabulary RGB-Infrared Benchmark for Multi-Angle Segmentation across Aerial and Ground Perspectives},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/M2326VRZ}},
  note         = {Machine review of arXiv:2506.03709}
}
read the original abstract

Open-vocabulary semantic segmentation (OVSS) involves assigning labels to each pixel in an image based on textual descriptions, leveraging world models like CLIP. However, they encounter significant challenges in cross-domain generalization, hindering their practical efficacy in real-world applications. Embodied AI systems are transforming autonomous navigation for ground vehicles and drones by enhancing their perception abilities, and in this study, we present AetherVision-Bench, a benchmark for multi-angle segmentation across aerial, and ground perspectives, which facilitates an extensive evaluation of performance across different viewing angles and sensor modalities. We assess state-of-the-art OVSS models on the proposed benchmark and investigate the key factors that impact the performance of zero-shot transfer models. Our work pioneers the creation of a robustness benchmark, offering valuable insights and establishing a foundation for future research.

Figures

Figures reproduced from arXiv: 2506.03709 by the authors.

Figure 1
Figure 1. Radar chart comparing test sets across AetherVision [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Comparison of viewpoints across three settings: (1) [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

38 extracted references · 23 canonical work pages

  1. [1]

    Syn2real domain generalization for underwater mine-like object de- tection using side-scan sonar

    Aayush Agrawal, Aniruddh Sikdar, Rajini Makam, Suresh Sundaram, Suresh Kumar Besai, and Mahesh Gopi. Syn2real domain generalization for underwater mine-like object de- tection using side-scan sonar. IEEE Geoscience and Remote Sensing Letters, 2025. 1

    work page 2025

  2. [2]

    What a mess: Multi-domain evaluation of zero-shot semantic segmentation

    Benedikt Blumenstiel, Johannes Jakubik, Hilde K ¨uhne, and Michael V¨ossing. What a mess: Multi-domain evaluation of zero-shot semantic segmentation. Advances in Neural Infor- mation Processing Systems, 36, 2024. 1, 2, 3

    work page 2024

  3. [3]

    Coco- stuff: Thing and stuff classes in context

    Holger Caesar, Jasper Uijlings, and Vittorio Ferrari. Coco- stuff: Thing and stuff classes in context. In Proceedings of the IEEE conference on computer vision and pattern recog- nition, pages 1209–1218, 2018. 3

    work page 2018

  4. [4]

    Cat- seg: Cost aggregation for open-vocabulary semantic seg- mentation

    Seokju Cho, Heeseong Shin, Sunghwan Hong, Anurag Arnab, Paul Hongsuck Seo, and Seungryong Kim. Cat- seg: Cost aggregation for open-vocabulary semantic seg- mentation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 4113– 4123, 2024. 1, 2, 3, 4

    2024

  5. [5]

    De- coupling zero-shot semantic segmentation

    Jian Ding, Nan Xue, Gui-Song Xia, and Dengxin Dai. De- coupling zero-shot semantic segmentation. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 11583–11592, 2022. 1, 3, 4

    work page 2022

  6. [6]

    Open- vocabulary panoptic segmentation maskclip

    Zheng Ding, Jieke Wang, and Zhuowen Tu. Open- vocabulary panoptic segmentation maskclip. 2022. 2

    2022

  7. [7]

    The pascal visual object classes challenge: A retrospective

    Mark Everingham, SM Ali Eslami, Luc Van Gool, Christo- pher KI Williams, John Winn, and Andrew Zisserman. The pascal visual object classes challenge: A retrospective. In- ternational journal of computer vision , 111:98–136, 2015. 3

    work page 2015

  8. [8]

    Scal- ing open-vocabulary image segmentation with image-level labels

    Golnaz Ghiasi, Xiuye Gu, Yin Cui, and Tsung-Yi Lin. Scal- ing open-vocabulary image segmentation with image-level labels. In European Conference on Computer Vision, pages 540–557. Springer, 2022. 1, 2, 3

    work page 2022

Show all 38 references
  1. [9]

    Multispectral video se- mantic segmentation: A benchmark dataset and baseline

    Wei Ji, Jingjing Li, Cheng Bian, Zongwei Zhou, Jiaying Zhao, Alan L Yuille, and Li Cheng. Multispectral video se- mantic segmentation: A benchmark dataset and baseline. In Proceedings of the IEEE/CVF Conference on Computer Vi- sion and Pattern Recognition, pages 1094–1104, 2023. 3

    2023

  2. [10]

    Scaling up visual and vision-language representa- tion learning with noisy text supervision

    Chao Jia, Yinfei Yang, Ye Xia, Yi-Ting Chen, Zarana Parekh, Hieu Pham, Quoc Le, Yun-Hsuan Sung, Zhen Li, and Tom Duerig. Scaling up visual and vision-language representa- tion learning with noisy text supervision. In International conference on machine learning, pages 4904–4916. PMLR,

  3. [11]

    An efficient approach with dynamic multiswarm of uavs for forest firefighting

    Josy John, K Harikumar, J Senthilnath, and Suresh Sun- daram. An efficient approach with dynamic multiswarm of uavs for forest firefighting. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 54(5):2860–2871, 2024. 1

    2024

  4. [12]

    A resource-efficient decentralized sequential planner for spa- tiotemporal wildfire mitigation

    Josy John, Shridhar Velhal, and Suresh Sundaram. A resource-efficient decentralized sequential planner for spa- tiotemporal wildfire mitigation. IEEE Transactions on Au- tomation Science and Engineering, 2025. 1

    2025

  5. [13]

    Fine-tuning can distort pretrained fea- tures and underperform out-of-distribution

    Ananya Kumar, Aditi Raghunathan, Robbie Jones, Tengyu Ma, and Percy Liang. Fine-tuning can distort pretrained fea- tures and underperform out-of-distribution. arXiv preprint arXiv:2202.10054, 2022. 1

    2022 arXiv

  6. [14]

    Caltech aerial rgb-thermal dataset in the wild

    Connor Lee, Matthew Anderson, Nikhil Ranganathan, Xingxing Zuo, Kevin Do, Georgia Gkioxari, and Soon-Jo Chung. Caltech aerial rgb-thermal dataset in the wild. In European Conference on Computer Vision, pages 236–256. Springer, 2024. 3

    2024

  7. [15]

    Open-vocabulary semantic segmentation with mask-adapted clip

    Feng Liang, Bichen Wu, Xiaoliang Dai, Kunpeng Li, Yinan Zhao, Hang Zhang, Peizhao Zhang, Peter Vajda, and Diana Marculescu. Open-vocabulary semantic segmentation with mask-adapted clip. In Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition, pag...

    2023

  8. [16]

    Holistic evalu- ation of language models

    Percy Liang, Rishi Bommasani, Tony Lee, Dimitris Tsipras, Dilara Soylu, Michihiro Yasunaga, Yian Zhang, Deepak Narayanan, Yuhuai Wu, Ananya Kumar, et al. Holistic evalu- ation of language models. arXiv preprint arXiv:2211.09110,

    arXiv

  9. [17]

    Uavid: A semantic segmentation dataset for uav imagery

    Ye Lyu, George V osselman, Gui-Song Xia, Alper Yilmaz, and Michael Ying Yang. Uavid: A semantic segmentation dataset for uav imagery. ISPRS journal of photogrammetry and remote sensing, 165:108–119, 2020. 3

    2020

  10. [18]

    Learning transferable visual models from natural language supervi- sion

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

    2021

  11. [19]

    Lca-on-the-line: Benchmarking out-of-distribution generalization with class taxonomies

    Jia Shi, Gautam Gare, Jinjin Tian, Siqi Chai, Zhiqiu Lin, Arun Vasudevan, Di Feng, Francesco Ferroni, and Shu Kong. Lca-on-the-line: Benchmarking out-of-distribution generalization with class taxonomies. arXiv preprint arXiv:2407.16067, 2024. 1

    2024 arXiv

  12. [20]

    Fully complex-valued fully con- volutional multi-feature fusion network (fc 2 mfn) for build- ing segmentation of insar images

    Aniruddh Sikdar, Sumanth Udupa, Suresh Sundaram, and Narasimhan Sundararajan. Fully complex-valued fully con- volutional multi-feature fusion network (fc 2 mfn) for build- ing segmentation of insar images. In 2022 IEEE Symposium Series on Computational Intelligence (SSCI) , pages 581–

    2022

  13. [21]

    Deepmao: Deep multi-scale aware over- complete network for building segmentation in satellite im- agery

    Aniruddh Sikdar, Sumanth Udupa, Prajwal Gurunath, and Suresh Sundaram. Deepmao: Deep multi-scale aware over- complete network for building segmentation in satellite im- agery. In Proceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition, pages 487–496...

    2023

  14. [22]

    Ssl-rgb2ir: Semi-supervised rgb-to-ir image-to-image translation for enhancing visual task train- ing in semantic segmentation and object detection

    Aniruddh Sikdar, Qiranul Saadiyean, Prahlad Anand, and Suresh Sundaram. Ssl-rgb2ir: Semi-supervised rgb-to-ir image-to-image translation for enhancing visual task train- ing in semantic segmentation and object detection. In 2024 IEEE/RSJ International Conference on Intelligent...

    2024

  15. [23]

    Skd- net: Spectral-based knowledge distillation in low-light ther- mal imagery for robotic perception

    Aniruddh Sikdar, Jayant Teotia, and Suresh Sundaram. Skd- net: Spectral-based knowledge distillation in low-light ther- mal imagery for robotic perception. In 2024 IEEE Inter- national Conference on Robotics and Automation (ICRA) , pages 9041–9047. IEEE, 2024. 2

    2024

  16. [24]

    Saga: Semantic-aware gray color augmen- tation for visible-to-thermal domain adaptation across multi- view drone and ground-based vision systems

    Aniruddh Sikdar, Prajwal Gurunath, Sumanth Udupa, Suresh Sundaram, et al. Saga: Semantic-aware gray color augmen- tation for visible-to-thermal domain adaptation across multi- view drone and ground-based vision systems. arXiv preprint arXiv:2504.15728, 2025. 1, 3

    2025 arXiv

  17. [25]

    Ogp- net: Optical guidance meets pixel-level contrastive distilla- tion for robust multi-modal and missing modality segmen- tation

    Aniruddh Sikdar, Jayant Teotia, and Suresh Sundaram. Ogp- net: Optical guidance meets pixel-level contrastive distilla- tion for robust multi-modal and missing modality segmen- tation. In Proceedings of the AAAI Conference on Artificial Intelligence, pages 6922–6930, 2025. 2

    2025

  18. [26]

    Isprs potsdam dataset within the isprs test project on urban classification, 3d building reconstruction and semantic labeling, 2012

    BSF Swissphoto. Isprs potsdam dataset within the isprs test project on urban classification, 3d building reconstruction and semantic labeling, 2012. 3

    2012

  19. [27]

    Piafusion: A progressive infrared and visible im- age fusion network based on illumination aware.Information Fusion, 83:79–92, 2022

    Linfeng Tang, Jiteng Yuan, Hao Zhang, Xingyu Jiang, and Jiayi Ma. Piafusion: A progressive infrared and visible im- age fusion network based on illumination aware.Information Fusion, 83:79–92, 2022. 3

    2022

  20. [28]

    Mrfp: Learning generalizable semantic segmentation from sim-2-real with multi-resolution feature perturbation

    Sumanth Udupa, Prajwal Gurunath, Aniruddh Sikdar, and Suresh Sundaram. Mrfp: Learning generalizable semantic segmentation from sim-2-real with multi-resolution feature perturbation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 59...

    2024

  21. [29]

    Robust fine-tuning of zero-shot models

    Mitchell Wortsman, Gabriel Ilharco, Jong Wook Kim, Mike Li, Simon Kornblith, Rebecca Roelofs, Raphael Gon- tijo Lopes, Hannaneh Hajishirzi, Ali Farhadi, Hongseok Namkoong, et al. Robust fine-tuning of zero-shot models. In Proceedings of the IEEE/CVF conference on computer vi- ...

    2022

  22. [30]

    Open-vocabulary panop- tic segmentation with text-to-image diffusion models

    Jiarui Xu, Sifei Liu, Arash Vahdat, Wonmin Byeon, Xiao- long Wang, and Shalini De Mello. Open-vocabulary panop- tic segmentation with text-to-image diffusion models. In Proceedings of the IEEE/CVF Conference on Computer Vi- sion and Pattern Recognition, pages 2955–2966, 2023. 2

    2023

  23. [31]

    A simple baseline for open- vocabulary semantic segmentation with pre-trained vision- language model

    Mengde Xu, Zheng Zhang, Fangyun Wei, Yutong Lin, Yue Cao, Han Hu, and Xiang Bai. A simple baseline for open- vocabulary semantic segmentation with pre-trained vision- language model. In European Conference on Computer Vi- sion, pages 736–753. Springer, 2022. 1, 3

    2022

  24. [32]

    Side adapter network for open-vocabulary semantic segmentation

    Mengde Xu, Zheng Zhang, Fangyun Wei, Han Hu, and Xi- ang Bai. Side adapter network for open-vocabulary semantic segmentation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 2945– 2954, 2023. 4

    2023

  25. [33]

    Convolutions die hard: Open-vocabulary seg- mentation with single frozen convolutional clip

    Qihang Yu, Ju He, Xueqing Deng, Xiaohui Shen, and Liang- Chieh Chen. Convolutions die hard: Open-vocabulary seg- mentation with single frozen convolutional clip. Advances in Neural Information Processing Systems, 36:32215–32234,

  26. [34]

    Florence: A new foundation model for computer vision

    Lu Yuan, Dongdong Chen, Yi-Ling Chen, Noel Codella, Xiyang Dai, Jianfeng Gao, Houdong Hu, Xuedong Huang, Boxin Li, Chunyuan Li, et al. Florence: A new foundation model for computer vision. arXiv preprint arXiv:2111.11432, 2021. 1

    2021 arXiv

  27. [35]

    Dept: Decoupled prompt tuning

    Ji Zhang, Shihan Wu, Lianli Gao, Heng Tao Shen, and Jingkuan Song. Dept: Decoupled prompt tuning. In Pro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 12924–12933, 2024. 1

    2024

  28. [36]

    Semantic under- standing of scenes through the ade20k dataset

    Bolei Zhou, Hang Zhao, Xavier Puig, Tete Xiao, Sanja Fi- dler, Adela Barriuso, and Antonio Torralba. Semantic under- standing of scenes through the ade20k dataset. International Journal of Computer Vision, 127:302–321, 2019. 3

    2019

  29. [37]

    Extract free dense labels from clip

    Chong Zhou, Chen Change Loy, and Bo Dai. Extract free dense labels from clip. In European Conference on Com- puter Vision, pages 696–712. Springer, 2022. 1

    2022

  30. [38]

    Generalized decoding for pixel, image, and lan- guage

    Xueyan Zou, Zi-Yi Dou, Jianwei Yang, Zhe Gan, Linjie Li, Chunyuan Li, Xiyang Dai, Harkirat Behl, Jianfeng Wang, Lu Yuan, et al. Generalized decoding for pixel, image, and lan- guage. In Proceedings of the IEEE/CVF conference on com- puter vision and pattern recognition , pages...

Pith tools

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