Pith. sign in

REVIEW 3 major objections 5 minor 2 cited by

VoCap: Video Object Captioning and Segmentation from Any Prompt

T0 review · 3 major / 5 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read The paper claims a single promptable video model can return both a pixel-level mask and a free-form caption for any prompted object.

desk verdict VoCap is a real contribution — the first promptable video model that outputs both masklets and captions, plus a large new pseudo-labeled dataset — but the captioning benchmark is small and rests on noisy Gemini labels, so those headline numbers should be read with caution. read the letter →

arxiv 2508.21809 v1 pith:3COBNVJT submitted 2025-08-29 cs.CV

Jasper Uijlings , Xingyi Zhou , Xiuye Gu , Arsha Nagrani , Anurag Arnab , Alireza Fathi , David Ross , Cordelia Schmid This is my paper
classification cs.CV
keywords videoobjectcaptioningreferringsegmentationsemi-supervisedpromptablemaskletpseudo-labelingvision-languagemodel
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

VoCap is a video model that takes one prompt—a text description, a bounding box, or a mask—and returns both a spatio-temporal masklet (a pixel-level mask that tracks the object across frames) and a free-form sentence describing the object. The paper argues that no previous system combined fine-grained video localization with open-ended object captioning, and that this combination can be trained at scale without expensive manual captions. To get training data, it starts from a large existing video-segmentation dataset and uses a large vision-language model to write pseudo-captions for 170k objects, then collects human-written captions on the validation set. On referring-expression video segmentation the model is reported to outperform prior methods, while remaining competitive on semi-supervised video segmentation and setting the first dedicated benchmark for video object captioning. If these results hold, the same weights can serve segmentation, tracking, and language description in one video understanding system.

What carries the argument

The architecture stacks a SAM2-style memory-based segmentation pipeline (image encoder, memory bank, mask decoder) with a lightweight BERT-style text encoder/decoder and a cross-attention caption feature extractor with learned query tokens. The prompt can be text, a first-frame box, or a mask; the same language module encodes an input text prompt and decodes the output caption. Data is generated by a pseudo-labeling pipeline: ground-truth masks from the SAV dataset are rendered as red contours with blurred backgrounds, and a large vision-language model writes structured captions (category, properties, actions). This pipeline converts an existing segmentation dataset into 170k masklet-caption

What would settle it

Take a held-out set of roughly 1,000 randomly selected SAV objects with human captions and compare VoCap's captions with the pseudo-labels and with a model trained without pseudo-captions. If the pseudo-trained model describes a nearby person or animal as often as the raw pseudo-labels do (the 'actor bias' reported in the paper), the claim that tracking plus pseudo-label training fixes the labeling noise would be contradicted. If a model trained without SAV-Caption data matches the captioning score on the manual validation set, the pseudo-caption pipeline is not the load-bearing component.

Watch

Extended reading notes

Core claim

The central claim is that one model can jointly solve promptable video object segmentation, referring-expression segmentation, and object captioning. Given a video and a prompt (text, box, or mask), VoCap predicts a masklet and a caption, using a memory-based tracker for temporal consistency and a shared text encoder/decoder for language. The paper also claims that pseudo-captions produced by a large vision-language model—shown the target object outlined in red with the background blurred—are good enough to train this model, and that co-training on the resulting 50k-video SAV-Caption dataset with existing image and video segmentation data improves both captioning and referring segmentation.

Load-bearing premise

The automatically generated training captions are accurate enough that a model trained on them generalizes to human annotations; the paper's own spot check found 12.0–15.5% of categories, properties, or actions wrong in a 50-video sample.

Editorial extensions

If this is right

  • One set of weights now covers promptable segmentation, referring-expression video object segmentation, and video object captioning, so a single model can replace task-specific pipelines.
  • Automatically generated captions from a large vision-language model, guided by visual prompts, are a viable substitute for manual video caption annotations at scale.
  • The new human-annotated SAV-Caption validation set provides the first dedicated evaluation benchmark for video object captioning.
  • Adding caption supervision also helps segmentation: removing the SAV-Caption training data lowers referring-segmentation performance, showing the two tasks reinforce each other.

Reading between the lines

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

  • Because the text encoder and decoder share weights, the model could in principle turn a caption back into a text prompt for segmentation, forming a closed loop for interactive video editing; the paper does not test this.
  • The pseudo-label noise documented in the appendix suggests a straightforward improvement: filter or re-rank generated captions (for example, by model confidence or cross-frame consistency) before training, which the paper does not do.
  • The same pseudo-labeling recipe could be applied to other mask-annotated video corpora to grow the training set further, likely improving caption diversity and reliability.
  • The SAV-Caption pairs (masklet + caption) could serve as pretraining data for video-language models beyond captioning, such as video question answering or text-driven video editing, since they ground language in precise spatio-temporal regions.
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 / 5 minor

Summary. The paper proposes VoCap, a unified model that takes a video and a prompt of any of three modalities (text, box, or mask) and outputs both a spatio-temporal masklet and a free-form object caption. The architecture extends SAM2-style memory-based segmentation with a shared BERT text encoder/decoder and a QFormer-like caption feature extractor. To obtain training data, the authors pseudo-label the SAV dataset (50k videos, 170k objects) using Gemini 1.5 Pro with red-contour highlighting and background blur, and they collect manual captions on SAV-val (155 videos, 290 objects, three captions per object) as a separate benchmark. The model is co-trained on VisualGenome, RefCOCO, RefVOS-YTVOS, and SAV-Caption-train, then finetuned per task. Experiments report video object captioning results on SAV-Caption-val, localized image captioning on VisualGenome, SS-VOS on YTVOS 2018 and MOSE, and RefVOS on RefVOS-DAVIS, RefVOS-YTVOS, MeViS, and UVO-VLN, claiming state-of-the-art RefVOS performance and competitive SS-VOS performance.

Significance. If the reported results are robust, VoCap is a useful contribution: it is, to my knowledge, the first promptable video model that jointly produces masklets and detailed object captions from text, box, or mask prompts, and it releases a large pseudo-labeled dataset plus a manually annotated validation benchmark. The experimental effort is substantial: several numbers come from official test servers, the method is compared against strong baselines including the authors' own SAM2 re-implementation and a Gemini pseudo-labeling pipeline, and the dataset and qualitative analyses are provided. The main risk is that the central captioning claim rests on a small validation set and on pseudo-label quality that is only weakly supported. The segmentation results, by contrast, are more solidly grounded in established benchmarks.

major comments (3)
  1. [§5.2, Table 3, and §A.1, Table 7] The headline video-object-captioning result is evaluated on only 155 videos / 290 objects with three references each, and Table 3 reports single CIDEr values without confidence intervals or significance tests. The 7.3-point gap over the SAM2+Gemini baseline (47.8 vs 40.5) may be within evaluation noise at this size. This is load-bearing because captioning is the paper's main novelty. The pseudo-label audit in Table 7 also reports 12.0% wrong categories, 12.4% wrong properties, and 15.5% wrong actions on only 50 objects, and Fig. 4 documents an actor-bias failure. Table 6 shows the pseudo-labels are useful for training, but usefulness does not establish label correctness; a model trained on Gemini's distribution could internalize the same biases and still beat a zero-shot Gemini baseline. Please provide bootstrapped CIs or per-object significance tests, a larger manual evaluation set, and
  2. [§5.3 and Table 5] There is an internal inconsistency in the SS-VOS numbers and in the accompanying claims. The text states 'our method yields 85.5 J&F' on YTVOS 2018, but Table 5 lists 85.0 for VoCap, equal to the re-implemented SAM2; on MOSE VoCap (66.3) is slightly below SAM2 (66.4). The conclusion's phrase 'state-of-the-art on video object segmentation' is therefore not supported by the table. Similarly, the abstract's 'state-of-the-art results on referring expression video object segmentation' should be qualified: without the FindTrack test-time procedure, VoCap is below GLEE on RefVOS-YTVOS (70.3 vs 70.6), and the SOTA claim holds only with FindTrack. The numbers and claims need to be reconciled.
  3. [§3.2 and §A.1] The manual annotation instructions mirror the structure of the pseudo-label prompt (category, properties, action). This ensures consistency between training and evaluation, but it also means the benchmark is tailored to the exact output format VoCap is trained to produce; it does not test truly free-form, unconstrained object captioning. The paper should acknowledge this as a scope limitation and report inter-annotator agreement or caption diversity statistics, which are currently missing. Since the claimed contribution includes 'free-form caption' output, this is a load-bearing qualification rather than a mere presentation issue.
minor comments (5)
  1. [Abstract and §5.3] The abstract/conclusion should consistently distinguish the online VoCap setting from VoCap+FindTrack, and the SS-VOS claim should be phrased as competitive rather than state-of-the-art, given the tie with SAM2 on YTVOS 2018.
  2. [§5.2] Typo: 'evalate' should be 'evaluate'. Also, the captioning baselines omit OW-VISCapTor and DenseVOC, which are discussed in Related Work; please clarify why they are not comparable for the mask-prompted protocol.
  3. [Figure 4] Caption contains 'succeesd' instead of 'succeeded'.
  4. [§3.1 / Table 1] The pseudo-caption pipeline is described as producing ~11.8 words per caption, but the human validation captions average 13.5 words. A brief discussion of this distributional difference would help readers interpret CIDEr comparisons.
  5. [§5.4 / Table 6] The 0% SAV-Caption row shows a large drop in SS-VOS J&F (57.7), which is interpreted as task synergy. It would be useful to state explicitly whether all other data ratios use the same number of training steps, since the absolute drops could partly reflect reduced video-data volume.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the central claims are empirical and are evaluated against independently collected human annotations and external benchmarks; the pseudo-label noise acknowledged in Appendix A.1 is a data-quality limitation, not a derivation-level loop.

full rationale

The paper makes no first-principles derivation claims; it is an empirical systems paper. The captioning claim is tested on SAV-Caption-val, where captions were manually collected by three annotators (Sec. 3.2), not generated by the same pseudo-label pipeline. The RefVOS and SS-VOS claims are tested on external benchmarks (RefVOS-YTVOS, MeViS, UVO-VLN, YTVOS, MOSE) via official test servers. Thus the reported predictions are not equivalent to training inputs by construction. Self-citations (e.g., PixelLLM [90] as a baseline, DenseVOC [109] for generic decoder design) are not load-bearing: no uniqueness theorem is invoked, and the cited architectural choices are also supported by external work (SAM2, BLIP2/QFormer). The paper's own quality audit (Appendix A.1, Table 7) reports 12.0% wrong categories, 12.4% wrong properties, and 15.5% wrong actions in a 50-object sample; this is an explicit data-quality caveat, not a circular step. One evaluation-design caveat is that the manual annotation instructions (Sec. 3.2) mirror the pseudo-label prompt structure (object class, visual properties, action), which may align the evaluation with the training label distribution; however, the manual references are human-written and the model is not fitted to them, so no result reduces to its input by definition.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The model is empirical; the central claim depends on standard training hyper-parameters as free parameters, the assumption that pseudo-labels are good, and the validity of the evaluation. No invented entities.

free parameters (3)
  • Multi-task data mixture ratio = 0.5:1:1:2 (VisualGenome:RefCOCO:RefVOS-YTVOS:SAV-Caption-train)
    Hand-chosen in Table 2 to balance datasets; affects relative task performance, and no sensitivity study is reported.
  • Input resolution = 512x512
    Chosen to fit hardware (Appendix B), differing from SAM2's 1024; the paper verifies minimal drop but it is a design choice.
  • Number of caption tokens = 32
    Taken from BLIP2 (Section 4.1) but still a hand-chosen architecture constant.
assumptions (5)
  • domain assumption SAV Manual masks are accurate enough to serve as ground truth for both pseudo-labelling and evaluation.
    The dataset pipeline builds on SAV Manual, which combines SAM2 predictions with human corrections (Section 3).
  • domain assumption Gemini 1.5 Pro Vision produces sufficiently accurate object-centric captions when guided by visual prompting.
    Pseudo-labels are the main training signal; quality is checked only on 50 objects (Appendix A.1).
  • domain assumption Manual caption annotations on SAV-val are an unbiased benchmark.
    They are collected independently, but instructions mirror the pseudo-label structure (class, properties, action), potentially biasing the metric (Section 3.2).
  • standard math The evaluation metrics (CIDEr, J&F) are computed correctly and are appropriate for the tasks.
    Standard metrics used in prior work; reliance on official test servers for most datasets.
  • domain assumption The model architecture (memory-based tracking + cross-attention captioning) can jointly learn mask and caption from multi-task data.
    This is the design hypothesis; supported by experiments, not provable.

how reviews work

0 comments
Cite this review

Pith. "Pith review of VoCap: Video Object Captioning and Segmentation from Any Prompt." pith.science (2026). https://pith.science/paper/3COBNVJT

@misc{pith2026250821809,
  author       = {Pith},
  title        = {Pith review of: VoCap: Video Object Captioning and Segmentation from Any Prompt},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3COBNVJT}},
  note         = {Machine review of arXiv:2508.21809}
}
read the original abstract

Understanding objects in videos in terms of fine-grained localization masks and detailed semantic properties is a fundamental task in video understanding. In this paper, we propose VoCap, a flexible video model that consumes a video and a prompt of various modalities (text, box or mask), and produces a spatio-temporal masklet with a corresponding object-centric caption. As such our model addresses simultaneously the tasks of promptable video object segmentation, referring expression segmentation, and object captioning. Since obtaining data for this task is tedious and expensive, we propose to annotate an existing large-scale segmentation dataset (SAV) with pseudo object captions. We do so by preprocessing videos with their ground-truth masks to highlight the object of interest and feed this to a large Vision Language Model (VLM). For an unbiased evaluation, we collect manual annotations on the validation set. We call the resulting dataset SAV-Caption. We train our VoCap model at scale on a SAV-Caption together with a mix of other image and video datasets. Our model yields state-of-the-art results on referring expression video object segmentation, is competitive on semi-supervised video object segmentation, and establishes a benchmark for video object captioning. Our dataset will be made available at https://github.com/google-deepmind/vocap.

Figures

Figures reproduced from arXiv: 2508.21809 by the authors.

Figure 1
Figure 1. Overview of our VoCap architecture. Our model processes videos frame-by-frame, with access to an updating memory for each object. Each frame goes through the image encoder, cross attends to the memory. The memory-aggregated image features and the object-specific prompt embeddings are fed into the mask decoder to obtain the mask predictions. The memory module is updated with the per-frame mask predictions and image f… view at source ↗
Figure 2
Figure 2. Illustration of our visual prompting. Top: the original frame; Bot￾tom: our processed input to the Gemini annotator. We apply a red contour to high￾light the target object and blur the back￾ground avoid distractions. A woman wearing a black puffer coat with fur trim on the hood and a polka dot dress is walking A barefoot person wearing a gray t-shirt and dark pants gets up from kneeling, picks up a cricket bat, and … view at source ↗
Figure 4
Figure 4. Qualitative examples which illustrate where VoCap succeesd where SAM2+Gemini pseudo [PITH_FULL_IMAGE:figures/full_fig_p018_4.png] view at source ↗

Discussion (0). Sign in to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Video Generation Models are General-Purpose Vision Learners

    cs.CV 2026-07 conditional novelty 6.0 of 10

    A video-diffusion backbone fine-tuned as a single-step multi-task perceiver matches or beats specialists on depth, normals, pose and segmentation, with high data efficiency and sim-to-real transfer.

  2. Watch, Remember, Reason: Human-View Video Understanding with MLLMs

    cs.CV 2026-06 unverdicted novelty 4.0 of 10

    This is a survey that frames video MLLM research via a human-view formulation of perceptual representations, memory states, reasoning traces, and predictions, then reviews methods, datasets, benchmarks, and open problems.

Reference graph

Works this paper leans on

110 extracted references · 54 canonical work pages · cited by 2 Pith papers

  1. [1]

    Flamingo: a visual language model for few-shot learning

    Jean-Baptiste Alayrac, Jeff Donahue, Pauline Luc, Antoine Miech, Iain Barr, Yana Hasson, Karel Lenc, Arthur Mensch, Katherine Millican, Malcolm Reynolds, et al. Flamingo: a visual language model for few-shot learning. NeurIPS, 2022

    2022

  2. [2]

    Frozen in time: A joint video and image encoder for end-to-end retrieval

    Max Bain, Arsha Nagrani, Gül Varol, and Andrew Zisserman. Frozen in time: A joint video and image encoder for end-to-end retrieval. In CVPR, 2021

    2021

  3. [3]

    Context r-cnn: Long term temporal context for per-camera object detection

    Sara Beery, Guanhang Wu, Vivek Rathod, Ronny V otel, and Jonathan Huang. Context r-cnn: Long term temporal context for per-camera object detection. In CVPR, 2020

    2020

  4. [4]

    JAX: composable transformations of Python+NumPy programs, 2018

    James Bradbury, Roy Frostig, Peter Hawkins, Matthew James Johnson, Chris Leary, Dougal Maclaurin, George Necula, Adam Paszke, Jake VanderPlas, Skye Wanderman-Milne, and Qiao Zhang. JAX: composable transformations of Python+NumPy programs, 2018

    2018

  5. [5]

    The 2019 davis challenge on vos: Unsupervised multi-object segmentation

    Sergi Caelles, Jordi Pont-Tuset, Federico Perazzi, Alberto Montes, Kevis-Kokitsi Maninis, and Luc Van Gool. The 2019 davis challenge on vos: Unsupervised multi-object segmentation. arXiv:1905.00737, 2019

    arXiv 2019

  6. [6]

    nuscenes: A multimodal dataset for autonomous driving

    Holger Caesar, Varun Bankiti, Alex H Lang, Sourabh V ora, Venice Erin Liong, Qiang Xu, Anush Krishnan, Yu Pan, Giancarlo Baldan, and Oscar Beijbom. nuscenes: A multimodal dataset for autonomous driving. In CVPR, 2020

    2020

  7. [7]

    Stablevideo: Text-driven consistency-aware diffusion video editing

    Wenhao Chai, Xun Guo, Gaoang Wang, and Yan Lu. Stablevideo: Text-driven consistency-aware diffusion video editing. In CVPR, 2023

    2023

  8. [8]

    Pali: A jointly-scaled multilingual language- image model

    Xi Chen, Xiao Wang, Soravit Changpinyo, AJ Piergiovanni, Piotr Padlewski, Daniel Salz, Sebastian Goodman, Adam Grycner, Basil Mustafa, Lucas Beyer, et al. Pali: A jointly-scaled multilingual language- image model. arXiv:2209.06794, 2022. 10

    arXiv 2022

Show all 110 references
  1. [9]

    Per-pixel classification is not all you need for semantic segmentation

    Bowen Cheng, Alex Schwing, and Alexander Kirillov. Per-pixel classification is not all you need for semantic segmentation. NeurIPS, 2021

    2021

  2. [10]

    Xmem: Long-term video object segmentation with an atkinson-shiffrin memory model

    Ho Kei Cheng and Alexander G Schwing. Xmem: Long-term video object segmentation with an atkinson-shiffrin memory model. In ECCV, 2022

    2022

  3. [11]

    Putting the object back into video object segmentation

    Ho Kei Cheng, Seoung Wug Oh, Brian Price, Joon-Young Lee, and Alexander Schwing. Putting the object back into video object segmentation. In CVPR, 2024

    2024

  4. [12]

    Segment and track anything

    Yangming Cheng, Liulei Li, Yuanyou Xu, Xiaodi Li, Zongxin Yang, Wenguan Wang, and Yi Yang. Segment and track anything. arXiv:2305.06558, 2023

    2023 arXiv

  5. [13]

    Find first, track next: Decoupling identification and propagation in referring video object segmentation

    Suhwan Cho, Seunghoon Lee, Minhyeok Lee, Jungho Lee, and Sangyoun Lee. Find first, track next: Decoupling identification and propagation in referring video object segmentation. In ArXiv, 2025

    2025

  6. [14]

    Ow-viscaptor: Abstractors for open-world video instance segmentation and captioning

    Anwesa Choudhuri, Girish Chowdhary, and Alexander G Schwing. Ow-viscaptor: Abstractors for open-world video instance segmentation and captioning. NeurIPS, 2024

    2024

  7. [15]

    Scenic: A jax library for computer vision research and beyond

    Mostafa Dehghani, Alexey Gritsenko, Anurag Arnab, Matthias Minderer, and Yi Tay. Scenic: A jax library for computer vision research and beyond. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 21393–21398, 2022

    2022

  8. [16]

    Memsam: Taming segment anything model for echocardiography video segmentation

    Xiaolong Deng, Huisi Wu, Runhao Zeng, and Jing Qin. Memsam: Taming segment anything model for echocardiography video segmentation. In CVPR, 2024

    2024

  9. [17]

    Bert: Pre-training of deep bidirectional transformers for language understanding

    Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. Bert: Pre-training of deep bidirectional transformers for language understanding. In NAACL, 2019

    2019

  10. [18]

    Mevis: A large-scale benchmark for video segmentation with motion expressions

    Henghui Ding, Chang Liu, Shuting He, Xudong Jiang, and Chen Change Loy. Mevis: A large-scale benchmark for video segmentation with motion expressions. In ICCV, 2023

    2023

  11. [19]

    MOSE: A new dataset for video object segmentation in complex scenes

    Henghui Ding, Chang Liu, Shuting He, Xudong Jiang, Philip HS Torr, and Song Bai. MOSE: A new dataset for video object segmentation in complex scenes. In ICCV, 2023

    2023

  12. [20]

    An image is worth 16x16 words: Transformers for image recognition at scale

    Alexey Dosovitskiy. An image is worth 16x16 words: Transformers for image recognition at scale. ICLR, 2021

    2021

  13. [21]

    Eva-02: A visual representation for neon genesis

    Yuxin Fang, Quan Sun, Xinggang Wang, Tiejun Huang, Xinlong Wang, and Yue Cao. Eva-02: A visual representation for neon genesis. arXiv:2303.11331, 2023

    2023 arXiv

  14. [22]

    Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context

    Google Gemini Team. Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context. arXiv:2403.05530, 2024

    2024 arXiv

  15. [23]

    Videosam: Open-world video segmentation

    Pinxue Guo, Zixu Zhao, Jianxiong Gao, Chongruo Wu, Tong He, Zheng Zhang, Tianjun Xiao, and Wenqiang Zhang. Videosam: Open-world video segmentation. arXiv:2410.08781, 2024

    2024 arXiv

  16. [24]

    Masked autoencoders are scalable vision learners

    Kaiming He, Xinlei Chen, Saining Xie, Yanghao Li, Piotr Dollár, and Ross Girshick. Masked autoencoders are scalable vision learners. In CVPR, 2022

    2022

  17. [25]

    Decoupling static and hierarchical motion perception for referring video segmentation

    Shuting He and Henghui Ding. Decoupling static and hierarchical motion perception for referring video segmentation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 13332–13341, 2024

    2024

  18. [26]

    Instruct-imagen: Image generation with multi-modal instruction

    Hexiang Hu, Kelvin CK Chan, Yu-Chuan Su, Wenhu Chen, Yandong Li, Kihyuk Sohn, Yang Zhao, Xue Ben, Boqing Gong, William Cohen, et al. Instruct-imagen: Image generation with multi-modal instruction. In CVPR, 2024

    2024

  19. [27]

    Segment and caption anything

    Xiaoke Huang, Jianfeng Wang, Yansong Tang, Zheng Zhang, Han Hu, Jiwen Lu, Lijuan Wang, and Zicheng Liu. Segment and caption anything. In CVPR, 2024

    2024

  20. [28]

    A better use of audio-visual cues: Dense video captioning with bi-modal transformer

    Vladimir Iashin and Esa Rahtu. A better use of audio-visual cues: Dense video captioning with bi-modal transformer. In BMVC, 2020

    2020

  21. [29]

    Perceiver: General perception with iterative attention

    Andrew Jaegle, Felix Gimeno, Andy Brock, Oriol Vinyals, Andrew Zisserman, and Joao Carreira. Perceiver: General perception with iterative attention. In ICML. PMLR, 2021

    2021

  22. [30]

    Scaling up visual and vision-language representation 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 representation learning with noisy text supervision. In ICML, 2021. 11

    2021

  23. [31]

    Densecap: Fully convolutional localization networks for dense captioning

    Justin Johnson, Andrej Karpathy, and Li Fei-Fei. Densecap: Fully convolutional localization networks for dense captioning. In CVPR, 2016

    2016

  24. [32]

    Kanani, Sriparna Saha, and Pushpak Bhattacharyya

    Chandresh S. Kanani, Sriparna Saha, and Pushpak Bhattacharyya. Global object proposals for improving multi-sentence video descriptions. In IJCNN, 2021

    2021

  25. [33]

    Video object segmentation with language referring expressions

    Anna Khoreva, Anna Rohrbach, and Bernt Schiele. Video object segmentation with language referring expressions. In ACCV, 2018

    2018

  26. [34]

    Video panoptic segmentation

    Dahun Kim, Sanghyun Woo, Joon-Young Lee, and In So Kweon. Video panoptic segmentation. InCVPR, 2020

    2020

  27. [35]

    Segment anything

    Alexander Kirillov, Eric Mintun, Nikhila Ravi, Hanzi Mao, Chloe Rolland, Laura Gustafson, Tete Xiao, Spencer Whitehead, Alexander C Berg, Wan-Yen Lo, et al. Segment anything. In ICCV, 2023

    2023

  28. [36]

    Dense-captioning events in videos

    Ranjay Krishna, Kenji Hata, Frederic Ren, Li Fei-Fei, and Juan Carlos Niebles. Dense-captioning events in videos. In ICCV, 2017

    2017

  29. [37]

    Visual genome: Connecting language and vision using crowdsourced dense image annotations

    Ranjay Krishna, Yuke Zhu, Oliver Groth, Justin Johnson, Kenji Hata, Joshua Kravitz, Stephanie Chen, Yannis Kalantidis, Li-Jia Li, David A Shamma, et al. Visual genome: Connecting language and vision using crowdsourced dense image annotations. IJCV, 2017

    2017

  30. [38]

    Referring video object segmentation with inter-frame interaction and cross-modal correlation

    Meng Lan, Fu Rong, and Lefei Zhang. Referring video object segmentation with inter-frame interaction and cross-modal correlation. arXiv:2307.00536, 2023

    2023 arXiv

  31. [39]

    Pseudo-label: The simple and efficient semi-supervised learning method for deep neural networks

    Dong-Hyun Lee. Pseudo-label: The simple and efficient semi-supervised learning method for deep neural networks. In ICML Workshop, 2013

    2013

  32. [40]

    Blip-2: Bootstrapping language-image pre- training with frozen image encoders and large language models

    Junnan Li, Dongxu Li, Silvio Savarese, and Steven Hoi. Blip-2: Bootstrapping language-image pre- training with frozen image encoders and large language models. In ICML, 2023

    2023

  33. [41]

    Learning object context for dense captioning

    Xiangyang Li, Shuqiang Jiang, and Jungong Han. Learning object context for dense captioning. In AAAI, 2019

    2019

  34. [42]

    Exploring plain vision transformer backbones for object detection

    Yanghao Li, Hanzi Mao, Ross Girshick, and Kaiming He. Exploring plain vision transformer backbones for object detection. In ECCV, 2022

    2022

  35. [43]

    Beyond mot: Semantic multi-object tracking

    Yunhao Li, Qin Li, Hao Wang, Xue Ma, Jiali Yao, Shaohua Dong, Heng Fan, and Libo Zhang. Beyond mot: Semantic multi-object tracking. In ECCV, 2024

    2024

  36. [44]

    Draw-and-understand: Leveraging visual prompts to enable mllms to comprehend what you want

    Weifeng Lin, Xinyu Wei, Ruichuan An, Peng Gao, Bocheng Zou, Yulin Luo, Siyuan Huang, Shanghang Zhang, and Hongsheng Li. Draw-and-understand: Leveraging visual prompts to enable mllms to comprehend what you want. arXiv:2403.20271, 2024

    2024 arXiv

  37. [45]

    Visual instruction tuning

    Haotian Liu, Chunyuan Li, Qingyang Wu, and Yong Jae Lee. Visual instruction tuning. In NeurIPS, 2023

    2023

  38. [46]

    Image segmentation using text and image prompts

    Timo Lüddecke and Alexander Ecker. Image segmentation using text and image prompts. In CVPR, 2022

    2022

  39. [47]

    Soc: Semantic-assisted object cluster for referring video object segmentation

    Zhuoyan Luo, Yicheng Xiao, Yong Liu, Shuyan Li, Yitong Wang, Yansong Tang, Xiu Li, and Yujiu Yang. Soc: Semantic-assisted object cluster for referring video object segmentation. NeurIPS, 2024

    2024

  40. [48]

    Generation and comprehension of unambiguous object descriptions

    Junhua Mao, Jonathan Huang, Alexander Toshev, Oana Camburu, Alan L Yuille, and Kevin Murphy. Generation and comprehension of unambiguous object descriptions. In CVPR, 2016

    2016

  41. [49]

    Scaling open-vocabulary object detection

    Matthias Minderer, Alexey Gritsenko, and Neil Houlsby. Scaling open-vocabulary object detection. NeurIPS, 2023

    2023

  42. [50]

    Pivot: Iterative visual prompting elicits actionable knowledge for vlms

    Soroush Nasiriany, Fei Xia, Wenhao Yu, Ted Xiao, Jacky Liang, Ishita Dasgupta, Annie Xie, Danny Driess, Ayzaan Wahid, Zhuo Xu, et al. Pivot: Iterative visual prompting elicits actionable knowledge for vlms. ICML, 2024

    2024

  43. [51]

    Gpt-4v(ision) technical work and authors

    OpenAI. Gpt-4v(ision) technical work and authors. https://cdn.openai.com/contributions/ gpt-4v.pdf, 2023,

    2023

  44. [52]

    Kosmos- 2: Grounding multimodal large language models to the world

    Zhiliang Peng, Wenhui Wang, Li Dong, Yaru Hao, Shaohan Huang, Shuming Ma, and Furu Wei. Kosmos- 2: Grounding multimodal large language models to the world. arXiv:2306.14824, 2023

    2023 arXiv

  45. [53]

    Perazzi, J

    F. Perazzi, J. Pont-Tuset, B. McWilliams, L. Van Gool, M. Gross, and A. Sorkine-Hornung. A benchmark dataset and evaluation methodology for video object segmentation. In CVPR, 2016. 12

    2016

  46. [54]

    Flickr30k entities: Collecting region-to-phrase correspondences for richer image-to-sentence models

    Bryan A Plummer, Liwei Wang, Chris M Cervantes, Juan C Caicedo, Julia Hockenmaier, and Svetlana Lazebnik. Flickr30k entities: Collecting region-to-phrase correspondences for richer image-to-sentence models. In CVPR, 2015

    2015

  47. [55]

    Connecting vision and language with localized narratives

    Jordi Pont-Tuset, Jasper Uijlings, Soravit Changpinyo, Radu Soricut, and Vittorio Ferrari. Connecting vision and language with localized narratives. In ECCV, 2020

    2020

  48. [56]

    Jiyang Qi, Yan Gao, Yao Hu, Xinggang Wang, Xiaoyu Liu, Xiang Bai, Serge Belongie, Alan Yuille, Philip H. S. Torr, and Song Bai. Occluded video instance segmentation: A benchmark. IJCV, 2022

    2022

  49. [57]

    Learning transferable visual models from natural language supervision

    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 supervision. In ICLR, 2021

    2021

  50. [58]

    Exploring the limits of transfer learning with a unified text-to-text transformer

    Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang, Michael Matena, Yanqi Zhou, Wei Li, and Peter J Liu. Exploring the limits of transfer learning with a unified text-to-text transformer. JMLR, 2020

    2020

  51. [59]

    Sam 2: Segment anything in images and videos

    Nikhila Ravi, Valentin Gabeur, Yuan-Ting Hu, Ronghang Hu, Chaitanya Ryali, Tengyu Ma, Haitham Khedr, Roman Rädle, Chloe Rolland, Laura Gustafson, et al. Sam 2: Segment anything in images and videos. arXiv:2408.00714, 2024

    2024 arXiv

  52. [60]

    Youtube- boundingboxes: A large high-precision human-annotated data set for object detection in video

    Esteban Real, Jonathon Shlens, Stefano Mazzocchi, Xin Pan, and Vincent Vanhoucke. Youtube- boundingboxes: A large high-precision human-annotated data set for object detection in video. In CVPR, 2017

    2017

  53. [61]

    U-net: Convolutional networks for biomedical image segmentation

    Olaf Ronneberger, Philipp Fischer, and Thomas Brox. U-net: Convolutional networks for biomedical image segmentation. In MICCAI, 2015

    2015

  54. [62]

    Berg, and Li Fei-Fei

    Olga Russakovsky, Jia Deng, Hao Su, Jonathan Krause, Sanjeev Satheesh, Sean Ma, Zhiheng Huang, Andrej Karpathy, Aditya Khosla, Michael Bernstein, Alexander C. Berg, and Li Fei-Fei. Imagenet large scale visual recognition challenge. IJCV, 2015

    2015

  55. [63]

    Hiera: A hierarchical vision transformer without the bells-and-whistles

    Chaitanya Ryali, Yuan-Ting Hu, Daniel Bolya, Chen Wei, Haoqi Fan, Po-Yao Huang, Vaibhav Aggarwal, Arkabandhu Chowdhury, Omid Poursaeed, Judy Hoffman, et al. Hiera: A hierarchical vision transformer without the bells-and-whistles. In ICLR, 2023

    2023

  56. [64]

    Tokenlearner: Adaptive space-time tokenization for videos

    Michael Ryoo, AJ Piergiovanni, Anurag Arnab, Mostafa Dehghani, and Anelia Angelova. Tokenlearner: Adaptive space-time tokenization for videos. NeurIPS, 2021

    2021

  57. [65]

    Urvos: Unified referring video object segmentation network with a large-scale benchmark

    Seonguk Seo, Joon-Young Lee, and Bohyung Han. Urvos: Unified referring video object segmentation network with a large-scale benchmark. In ECCV, 2020

    2020

  58. [66]

    Annotating objects and relations in user-generated videos

    Xindi Shang, Donglin Di, Junbin Xiao, Yu Cao, Xun Yang, and Tat-Seng Chua. Annotating objects and relations in user-generated videos. In ICMR, 2019

    2019

  59. [67]

    Region-object relation-aware dense captioning via transformer

    Zhuang Shao, Jungong Han, Demetris Marnerides, and Kurt Debattista. Region-object relation-aware dense captioning via transformer. IEEE Transactions on Neural Networks and Learning Systems , 2022

    2022

  60. [68]

    What does clip know about a red circle? visual prompt engineering for vlms

    Aleksandar Shtedritski, Christian Rupprecht, and Andrea Vedaldi. What does clip know about a red circle? visual prompt engineering for vlms. In ICCV, 2023

    2023

  61. [69]

    Video foundation models for animal behavior analysis

    Jennifer J Sun, Hao Zhou, Long Zhao, Liangzhe Yuan, Bryan Seybold, David Hendon, Florian Schroff, David A Ross, Hartwig Adam, Bo Hu, et al. Video foundation models for animal behavior analysis. bioRxiv, 2024

    2024

  62. [70]

    Scalability in perception for autonomous driving: Waymo open dataset

    Pei Sun, Henrik Kretzschmar, Xerxes Dotiwalla, Aurelien Chouard, Vijaysai Patnaik, Paul Tsui, James Guo, Yin Zhou, Yuning Chai, Benjamin Caine, et al. Scalability in perception for autonomous driving: Waymo open dataset. In CVPR, 2020

    2020

  63. [71]

    Gemma: Open models based on gemini research and technology

    Gemma Team, Thomas Mesnard, Cassidy Hardin, Robert Dadashi, Surya Bhupatiraju, Shreya Pathak, Laurent Sifre, Morgane Rivière, Mihir Sanjay Kale, Juliette Love, et al. Gemma: Open models based on gemini research and technology. arXiv:2403.08295, 2024

    2024 arXiv

  64. [72]

    Gemma 2: Improving open language models at a practical size

    Gemma Team, Morgane Riviere, Shreya Pathak, Pier Giuseppe Sessa, Cassidy Hardin, Surya Bhupatiraju, Léonard Hussenot, Thomas Mesnard, Bobak Shahriari, Alexandre Ramé, et al. Gemma 2: Improving open language models at a practical size. arXiv:2408.00118, 2024. 13

    2024 arXiv

  65. [73]

    Llama: Open and efficient foundation language models

    Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timothée Lacroix, Baptiste Rozière, Naman Goyal, Eric Hambro, Faisal Azhar, et al. Llama: Open and efficient foundation language models. arXiv:2302.13971, 2023

    2023 arXiv

  66. [74]

    Attention is all you need

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. Attention is all you need. In NeurIPS, 2017

    2017

  67. [75]

    Cider: Consensus-based image description evaluation

    Ramakrishna Vedantam, C Lawrence Zitnick, and Devi Parikh. Cider: Consensus-based image description evaluation. In CVPR, 2015

    2015

  68. [76]

    Phenaki: Variable length video generation from open domain textual descriptions

    Ruben Villegas, Mohammad Babaeizadeh, Pieter-Jan Kindermans, Hernan Moraldo, Han Zhang, Moham- mad Taghi Saffar, Santiago Castro, Julius Kunze, and Dumitru Erhan. Phenaki: Variable length video generation from open domain textual descriptions. In ICLR, 2022

    2022

  69. [77]

    Connecting vision and language with video localized narratives

    Paul V oigtlaender, Soravit Changpinyo, Jordi Pont-Tuset, Radu Soricut, and Vittorio Ferrari. Connecting vision and language with video localized narratives. In CVPR, 2023

    2023

  70. [78]

    Git: A generative image-to-text transformer for vision and language

    Jianfeng Wang, Zhengyuan Yang, Xiaowei Hu, Linjie Li, Kevin Lin, Zhe Gan, Zicheng Liu, Ce Liu, and Lijuan Wang. Git: A generative image-to-text transformer for vision and language. TMLR, 2022

    2022

  71. [79]

    End-to-end dense video captioning with parallel decoding

    Teng Wang, Ruimao Zhang, Zhichao Lu, Feng Zheng, Ran Cheng, and Ping Luo. End-to-end dense video captioning with parallel decoding. In ICCV, 2021

    2021

  72. [80]

    Caption anything: Interactive image description with diverse multimodal controls

    Teng Wang, Jinrui Zhang, Junjie Fei, Yixiao Ge, Hao Zheng, Yunlong Tang, Zhe Li, Mingqi Gao, Shanshan Zhao, Ying Shan, and Feng Zheng. Caption anything: Interactive image description with diverse multimodal controls. arXiv:2305.02677, 2023

    2023 arXiv

  73. [81]

    Unidentified video objects: A benchmark for dense, open-world segmentation

    Weiyao Wang, Matt Feiszli, Heng Wang, and Du Tran. Unidentified video objects: A benchmark for dense, open-world segmentation. In CVPR, 2021

    2021

  74. [82]

    The all-seeing project: Towards panoptic visual recognition and understanding of the open world

    Weiyun Wang, Min Shi, Qingyun Li, Wenhai Wang, Zhenhang Huang, Linjie Xing, Zhe Chen, Hao Li, Xizhou Zhu, Zhiguo Cao, et al. The all-seeing project: Towards panoptic visual recognition and understanding of the open world. ICLR, 2023

    2023

  75. [83]

    The all-seeing project v2: Towards general relation compre- hension of the open world

    Weiyun Wang, Yiming Ren, Haowen Luo, Tiantong Li, Chenxiang Yan, Zhe Chen, Wenhai Wang, Qingyun Li, Lewei Lu, Xizhou Zhu, et al. The all-seeing project v2: Towards general relation compre- hension of the open world. arXiv:2402.19474, 2024

    2024 arXiv

  76. [84]

    Instancediffusion: Instance-level control for image generation

    Xudong Wang, Trevor Darrell, Sai Saketh Rambhatla, Rohit Girdhar, and Ishan Misra. Instancediffusion: Instance-level control for image generation. In CVPR, 2024

    2024

  77. [85]

    Language as queries for referring video object segmentation

    Jiannan Wu, Yi Jiang, Peize Sun, Zehuan Yuan, and Ping Luo. Language as queries for referring video object segmentation. In CVPR, 2022

    2022

  78. [86]

    Uniref++: Segment every reference object in spatial and temporal spaces

    Jiannan Wu, Yi Jiang, Bin Yan, Huchuan Lu, Zehuan Yuan, and Ping Luo. Uniref++: Segment every reference object in spatial and temporal spaces. arXiv:2312.15715, 2023

    2023 arXiv

  79. [87]

    General object foundation model for images and videos at scale

    Junfeng Wu, Yi Jiang, Qihao Liu, Zehuan Yuan, Xiang Bai, and Song Bai. General object foundation model for images and videos at scale. In CVPR, 2024

    2024

  80. [88]

    Grit: A generative region-to-text transformer for object understanding

    Jialian Wu, Jianfeng Wang, Zhengyuan Yang, Zhe Gan, Zicheng Liu, Junsong Yuan, and Lijuan Wang. Grit: A generative region-to-text transformer for object understanding. ECCV, 2024

    2024

  81. [89]

    Dettoolchain: A new prompting paradigm to unleash detection ability of mllm

    Yixuan Wu, Yizhou Wang, Shixiang Tang, Wenhao Wu, Tong He, Wanli Ouyang, Philip Torr, and Jian Wu. Dettoolchain: A new prompting paradigm to unleash detection ability of mllm. ECCV, 2024

    2024

  82. [90]

    Pixel-aligned language model

    Jiarui Xu, Xingyi Zhou, Shen Yan, Xiuye Gu, Anurag Arnab, Chen Sun, Xiaolong Wang, and Cordelia Schmid. Pixel-aligned language model. In CVPR, 2024

    2024

  83. [91]

    Youtube-vos: Sequence-to-sequence video object segmentation

    Ning Xu, Linjie Yang, Yuchen Fan, Jianchao Yang, Dingcheng Yue, Yuchen Liang, Brian Price, Scott Cohen, and Thomas Huang. Youtube-vos: Sequence-to-sequence video object segmentation. In ECCV, 2018

    2018

  84. [92]

    xgen-mm (blip-3): A family of open large multimodal models

    Le Xue, Manli Shu, Anas Awadalla, Jun Wang, An Yan, Senthil Purushwalkam, Honglu Zhou, Viraj Prabhu, Yutong Dai, Michael S Ryoo, et al. xgen-mm (blip-3): A family of open large multimodal models. arXiv preprint arXiv:2408.08872, 2024

    2024

  85. [93]

    Vid2seq: Large-scale pretraining of a visual language model for dense video captioning

    Antoine Yang, Arsha Nagrani, Paul Hongsuck Seo, Antoine Miech, Jordi Pont-Tuset, Ivan Laptev, Josef Sivic, and Cordelia Schmid. Vid2seq: Large-scale pretraining of a visual language model for dense video captioning. In CVPR, 2023. 14

    2023

  86. [94]

    Track anything: Segment anything meets videos

    Jinyu Yang, Mingqi Gao, Zhe Li, Shang Gao, Fangjing Wang, and Feng Zheng. Track anything: Segment anything meets videos. arXiv:2304.11968, 2023

    2023 arXiv

  87. [95]

    Set-of-mark prompting unleashes extraordinary visual grounding in gpt-4v

    Jianwei Yang, Hao Zhang, Feng Li, Xueyan Zou, Chunyuan Li, and Jianfeng Gao. Set-of-mark prompting unleashes extraordinary visual grounding in gpt-4v. arXiv preprint arXiv:2310.11441, 2023

    2023 arXiv

  88. [96]

    Decoupling features in hierarchical propagation for video object segmentation

    Zongxin Yang and Yi Yang. Decoupling features in hierarchical propagation for video object segmentation. In NeurIPS, 2022

    2022

  89. [97]

    Associating objects with transformers for video object segmentation

    Zongxin Yang, Yunchao Wei, and Yi Yang. Associating objects with transformers for video object segmentation. In NeurIPS, 2021

    2021

  90. [98]

    Scalable video object segmentation with identification mechanism

    Zongxin Yang, Jiaxu Miao, Yunchao Wei, Wenguan Wang, Xiaohan Wang, and Yi Yang. Scalable video object segmentation with identification mechanism. TPAMI, 2024

    2024

  91. [99]

    Describing videos by exploiting temporal structure

    Li Yao, Atousa Torabi, Kyunghyun Cho, Nicolas Ballas, Christopher Pal, Hugo Larochelle, and Aaron Courville. Describing videos by exploiting temporal structure. In ICCV, 2015

    2015

  92. [100]

    Modeling context in referring expressions

    Licheng Yu, Patrick Poirson, Shan Yang, Alexander C Berg, and Tamara L Berg. Modeling context in referring expressions. In ECCV, 2016

    2016

  93. [101]

    Osprey: Pixel understanding with visual instruction tuning

    Yuqian Yuan, Wentong Li, Jian Liu, Dongqi Tang, Xinjie Luo, Chi Qin, Lei Zhang, and Jianke Zhu. Osprey: Pixel understanding with visual instruction tuning. In CVPR, 2024

    2024

  94. [102]

    Gpt4roi: Instruction tuning large language model on region-of-interest

    Shilong Zhang, Peize Sun, Shoufa Chen, Min Xiao, Wenqi Shao, Wenwei Zhang, Yu Liu, Kai Chen, and Ping Luo. Gpt4roi: Instruction tuning large language model on region-of-interest. arXiv:2307.03601, 2023

    2023 arXiv

  95. [103]

    Where does it exist: Spatio-temporal video grounding for multi-form sentences

    Zhu Zhang, Zhou Zhao, Yang Zhao, Qi Wang, Huasheng Liu, and Lianli Gao. Where does it exist: Spatio-temporal video grounding for multi-form sentences. In CVPR, 2020

    2020

  96. [104]

    Gpt-4v (ision) is a generalist web agent, if grounded

    Boyuan Zheng, Boyu Gou, Jihyung Kil, Huan Sun, and Yu Su. Gpt-4v (ision) is a generalist web agent, if grounded. ICML, 2024

    2024

  97. [105]

    Rethinking semantic segmentation from a sequence-to-sequence perspective with transformers

    Sixiao Zheng, Jiachen Lu, Hengshuang Zhao, Xiatian Zhu, Zekun Luo, Yabiao Wang, Yanwei Fu, Jianfeng Feng, Tao Xiang, Philip HS Torr, et al. Rethinking semantic segmentation from a sequence-to-sequence perspective with transformers. In CVPR, 2021

    2021

  98. [106]

    Weakly-supervised video object grounding from text by loss weighting and object interaction

    Luowei Zhou, Nathan Louis, and Jason J Corso. Weakly-supervised video object grounding from text by loss weighting and object interaction. In BMVC, 2018

    2018

  99. [107]

    Towards automatic learning of procedures from web instructional videos

    Luowei Zhou, Chenliang Xu, and Jason J Corso. Towards automatic learning of procedures from web instructional videos. In AAAI, 2018

    2018

  100. [108]

    Grounded video description

    Luowei Zhou, Yannis Kalantidis, Xinlei Chen, Jason J Corso, and Marcus Rohrbach. Grounded video description. In CVPR, 2019

    2019

  101. [109]

    Dense video object captioning from disjoint supervision

    Xingyi Zhou, Anurag Arnab, Chen Sun, and Cordelia Schmid. Dense video object captioning from disjoint supervision. arXiv:2306.11729, 2023

    2023 arXiv

  102. [110]

    Point-VOS: Pointing up video object segmentation

    Idil Esen Zulfikar, Sabarinath Mahadevan, Paul V oigtlaender, and Bastian Leibe. Point-VOS: Pointing up video object segmentation. In CVPR, 2024. 15 A Details on Dataset A.1 Quality of SA V-Caption-train We performed a quantitative evaluation on the quality of the SA V-Caption...

    2024

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.