Pith.

Computer Vision and Pattern Recognition · arXiv:2611.16417 v1 · submitted 12 Nov 2026 · 31 references

VBot Exo2Ego Policy: Lightweight Data Collection and Exocentric-to-Egocentric Video Generation

Tong Rao, Zhihao Zhang, Yuntian Wang, Siyu Zhou, Hailong Qin, Yinan Yu

The paper says

“We present VBot Exo2Ego Policy, a video generation model that turns third-person videos into first-person observations of a selected actor, offering a practical way to expand first-person operation data from abundant real-world footage.”

more of the abstract

To support this goal, we introduce a lightweight data collection pipeline and mobile app that support first-person recording, third-person recording, and synchronized capture of both views. This setup combines accessible single-view collection with paired RGB and camera geometry for learning cross-view correspondence. Building on a pretrained video diffusion transformer, our model incorporates geometric guidance and auxiliary supervision from semantic masks and hand keypoints to facilitate optimization and improve the fidelity of hand-object interactions. Camera geometry is injected stochastically during training and gradually removed, allowing inference without target-camera calibration. To accommodate the diversity of Internet videos, we combine source-video perception and reasoning, optional actor references, and multi-shot adaptation to identify the intended actor and follow an operation across camera cuts. The collection system and generation model provide a foundation for expanding first-person operation data across tasks, scenes, and participants.

from the abstract

Pith says

Machine review · deepseek-v4-flash · 13 Nov 2026 · what was read was not recorded · not checked by a person

It can be wrong. Challenge this review or reply to it.

Holds up, with conditions. Solid engineering with credible gains over EgoX, but the camera conditioning claim lacks a controlled ablation on the main evaluation set—the only evidence is descriptive. This line is wrong

3 major objections · 5 minor

How often objections from reviews like this one are right: not yet measured

Are you an author? What you can do.

Not checked: figures, code, what the 31 references say.

The paper says, at Section 6.4, Table 7, Appendix A.2

The review did not quote this passage.

A classifier read this sentence of the paper as acknowledging the concern (probability 0.66):

“The historical RGB reference uses a different data pool and learning-rate schedule, so its loss comparison is descriptive.”

Pith says: Objection 1 of 3 · major

The paper claims that stochastic camera conditioning enables training with calibrated rays while allowing uncalibrated inference (Section 4.2). However, the only direct evidence for this claim (Table 7) uses a different data pool and the paper itself states 'The historical RGB reference uses a different data pool and learning-rate schedule, so its loss comparison is descriptive' (Appendix A.2, sentence b338).

more of objection 1

No controlled ablation on the main 546-clip 448×448 evaluation set is performed. Without such an ablation, the contribution of ray conditioning to the headline PSNR/FVD gains (Table 3) is unknown. The authors should either run a proper ablation (e.g., removing the ray branch) on the main evaluation set, or clearly delimit the scope of the camera conditioning claim and acknowledge that its benefit has not been validated on the primary protocol.

The paper says, at Table 3, comparison details

The review did not quote this passage.

Pith says: Objection 2 of 3 · major

The baseline comparisons use different input frame counts (EgoX: 49 frames, EgoX/H3: 141 frames) and different base architectures (EgoX vs. MiniMax-H3). While a 49-frame version of VBot is reported, the gains may partially originate from the larger frame count or the stronger base model rather than the proposed components. A more controlled baseline—e.g., an identical base model trained without camera conditioning, semantic supervision, and hand supervision—would help attribute the improvements. The discussion should explicitly address the confounding factors.

The paper says, at Section 3, data collection pipeline

The review did not quote this passage.

Pith says: Objection 3 of 3 · major

The paper describes a lightweight data collection pipeline but provides no quantitative evaluation of its quality or efficiency: e.g., success rate of the common-start SOP and Sim(3) alignment, user effort, or comparison to existing datasets like Ego-Exo4D. Without such data, the 'lightweight' claim is unsupported. The authors should report statistics on alignment accuracy, data yield per recording session, and any quality control metrics.

5 minor objections

The paper says, at Section 3.2

Pith says: minor objection 1 of 5

The common-start SOP and joint reconstruction are described, but the success rate of the Pi3/DA3 reconstruction and Sim(3) alignment is not reported. How often does the procedure produce usable geometric pairing?

The paper says, at Equation (6), Section 4.2

Pith says: minor objection 2 of 5

The annealing schedule for ray probability uses a piecewise linear decay. Was this schedule tuned? What motivated the initial probability of 0.5? A brief justification would improve clarity.

show 3 more

The paper says, at Appendix A.2, Figure 5

Pith says: minor objection 3 of 5

The training loss curves in Figure 5 are informative, but the figure caption and labeling could be clearer: indicate which curve corresponds to stochastic rays and which to historical RGB, and note that the historical RGB uses a different data pool.

The paper says, at Section 5.2

Pith says: minor objection 4 of 5

The variants 'ACEaux' and 'MultiLayer' are used throughout Section 6.3 before being fully defined. Consider introducing them explicitly in Section 5.2 with a brief summary of their differences.

The paper says, at General

Pith says: minor objection 5 of 5

Minor typos and formatting inconsistencies: e.g., 'Sectio' in a few places (check references), variable capitalization of 'Exo/Ego' in figure captions. Also, the paper mentions 'checkpoint sensitivity' (Appendix A.3) but does not quantify variability; a brief analysis of training seed variance would strengthen the reliability claims.

What would settle it

Train two versions of VBot Exo2Ego Policy on the same paired dataset (main 546-clip evaluation set) with identical training schedule, hyperparameters, and loss weights: one with the stochastic ray conditioning branch (Bernoulli gate and annealing per Eq. 6) and one with the ray branch entirely removed throughout training. Evaluate both on the fixed 546-clip panel (Table 3). If ΔPSNR < 0.5 dB and ΔFVD < 10% on the unseen subset, the ray conditioning contributes negligibly to the headline results and the descriptive claim is unsupported; larger differences would validate the conditioning approach.

from the skeptic's reading

How the review went

4 readings by deepseek-v4-flash · 13 Nov 2026

  1. Reader holds up, with conditions confidence high
  2. Skeptic agrees 26 s · 19,541 tokens read, 3,190 written
  3. Editor wrote the verdict above 7 s · 20,446 tokens read, 782 written
  4. Referee major revision: 3 major objections, 5 minor 24 s · 19,969 tokens read, 3,032 written

Not run on this paper: the simulated authors' defence.

Pith did not check this review's objections against the paper.

The review in full

The review in brief, machine-written

The paper builds on known components (MiniMax-H3, ACE, Qwen) and adds stochastic camera conditioning with annealing, semantic HOI supervision, and hand keypoint supervision. The practical data collection pipeline is a real contribution, and the ablations for hand and semantic supervision are thorough and credible. The quantitative gains over EgoX (3.2 dB PSNR, ~38 % FVD reduction) are convincing as a system-level result, even if built on an upgraded backbone.

The soft spot is the camera conditioning claim. The paper admits (sentence b338) that the ray-conditioning ablation (Table 7) compares a model trained on a different data pool with a different schedule—i.e., it is descriptive, not controlled. That means the contribution of ray conditioning to the headline numbers is not empirically separable from the architecture, semantic supervision, or data. The stress-test note is correct: Table 7 does not use the main 546-clip 448×448 evaluation set, and no ablation removes the ray branch on that set. The authors could easily run a controlled ablation (train without rays on the same data), and I think a reviewer should ask for it. For now, the paper overclaims on what the camera module achieves.

Other concerns: reproducibility is low (no code/data released), and the checkpoint sensitivity noted in the appendix (FVD jumps at different steps) deserves acknowledgment in the main text. The semantic-mask and hand supervision results, however, stand on cleaner ablations.

Who is this for? Researchers working on view synthesis for manipulation data or EXO-to-EGO generation. It moves the needle incrementally over EgoX and provides a useful reference system.

Recommendation: Accept with major revision. The authors should add a controlled camera-conditioning ablation on the main evaluation set and release at least the model weights and a subset of paired data. The descriptive nature of Table 7 is a real weakness, but the rest of the paper is solid enough to warrant referee time.

The strongest objection

The paper's central claim—that stochastic camera conditioning enables training with calibrated rays yet allows uncalibrated inference—is not validated on the same evaluation protocol used for the headline comparison (Table 3). The sole supporting experiment (Table 7, Appendix A.6) compares flow-MSE between a model trained with ray conditioning on 33,056 clips at native 256P and a historical RGB reference trained on a different data pool and with a different schedule. The paper itself states 'The historical RGB reference uses a different data pool and learning-rate schedule, so its loss comparison is descriptive' (Appendix A.2, sentence b338). This means the contribution of ray conditioning to the reported PSNR/FVD gains of 3.23–3.25 dB and 37.6–39.8% is not empirically separable from other components (architecture, semantic/hand supervision, training data). The baselines (EgoX, EgoX/H3) use target-view geometric priors while VBot Exo2Ego Policy does not, yet no controlled ablation removes the ray branch on the main 546-clip 448×448 evaluation set to measure its effect. Without this, the claim that camera conditioning is beneficial generalizing to uncalibrated inference is weakly supported.

The reader identified the camera conditioning claim as the weakest assumption, noting the loss comparison is descriptive and uses a different data pool. My stress-test confirms this as the most load-bearing concern: if the ray conditioning does not actually contribute to the reported gains, then the paper's argument for a key design choice (stochastic injection with annealing) collapses. The reader's conditional verdict already calls for a more rigorous camera-conditioning ablation, which is exactly what the concrete test provides. I do not see a more fundamental flaw—other comparisons (semantic, hand supervision) are properly ablated on the main set, and the baseline comparisons are credible given identical clip sets and metrics. Therefore the verdict should remain unchanged: conditional acceptance hinges on the proposed controlled ablation, consistent with the reader's original condition.

The referee's report

If the results are reproducible, the work offers a practical pipeline for expanding egocentric manipulation data from more abundant exocentric sources, which is valuable for robotics, activity understanding, and AR/VR. The data collection standard and the model's ability to handle multi-person scenes and camera cuts address realistic deployment scenarios. The system-level gains over existing methods are substantial, though the specific contributions of each component need clearer validation.

Pith's reading of the paper (a machine's summary, not the authors')

The paper aims to establish that exocentric-to-egocentric video generation for manipulation tasks can be made practical through a lightweight data collection pipeline and a diffusion model with stochastic camera conditioning. If successful, anyone with a phone could record a task from a third-person view and automatically obtain a first-person video of the same action, vastly expanding training data for robotic or assistance systems. The model uses a pretrained video diffusion transformer (MiniMax-H3) trained with camera rays that are gradually removed, so at inference no calibration is needed. Auxiliary supervision on hand keypoints and object masks improves the fidelity of hand-object interactions. Quantitative evaluations show clear gains over prior work EgoX: PSNR +3.23 dB, FVD -37.6%.

What carries the argument. Stochastic camera conditioning adds Plücker rays to the EGO tokens during training via zero-initialized projections, with a Bernoulli gate that is annealed to zero so the final model does not depend on calibrated rays. Auxiliary supervision uses semantic masks for hand/object regions and ACE-Ego-Hand features plus 2D keypoints to shape the EGO representation for better hand articulation.

Core claim. VBot Exo2Ego Policy introduces a capture protocol that pairs two iPhones for synchronized ego/exo RGB-D video with ARKit poses, and a single-view crowdsourcing route for broader coverage. The generative model, built on a frozen video diffusion transformer, learns viewpoint conversion via stochastic Plücker-ray conditioning that is annealed away during training, allowing inference to proceed from just the exo video, text description, and an optional actor reference. Semantic hand/object masks and ACE-based hand feature supervision train the generator to produce videos from which hand joints and object masks can be reliably extracted by off-the-shelf models. The system outperforms EgoX on all evaluated metrics, with particular gains in hand localization accuracy.

The premise it rests on. The paired capture pipeline assumes that the common-start SOP provides sufficient visual overlap for reconstructing metric camera geometry, and that this geometry, learned stochastically, generalizes to uncalibrated inference after annealing.

If the paper is right (Pith's inferences, not the authors')

Large-scale generation of first-person operation videos from internet exo footage, enabling cheaper data expansion for robot learning or skill instruction.

The capture standard (paired iPhones plus common-start SOP) can be adopted by other labs to create high-quality ego-exo benchmarks for novel-view synthesis.

The stochastic conditioning method may generalize to other novel-view synthesis tasks where inference-time camera calibration is inconvenient or unavailable.

Hand supervision from ACE features demonstrates that diffusion generators can be trained to produce videos from which downstream perception modules recover more consistent joint structure.

The common-start SOP could eventually be replaced by automatic wide-baseline matching from feed-forward geometry models, making the data pipeline even lighter.

The annealing schedule might be replaced by consistency distillation or adversarial techniques to remove calibration requirements more directly.

This approach suggests that diffusion models can internalize 3D structure through stochastic geometric conditioning, opening the door to geometry-free inference for many view-conversion tasks.

If the model generalizes well, it could reduce the need for costly first-person data collection in dexterous manipulation learning across labs.

Check this review

Pith stored no record of this review's passes and took no fingerprint of it. Reviews made by code from 13 Nov 2026 on carry both.

How often reviews like this one are right

Reviews like this one: written by deepseek-v4-flash, where what was read was not recorded.

How often objections from reviews like this one are right: not yet measured

Every value above, in the review's public record (JSON)

Reviews by people

This review was written by a machine. No person has reviewed this paper on Pith yet. A scientist listed on Pith can review it for a fee, and their review is published here whatever it concludes.

Request a human review

How Pith checks papers · The full audit of this paper

The paper

Abstract

We present VBot Exo2Ego Policy, a video generation model that turns third-person videos into first-person observations of a selected actor, offering a practical way to expand first-person operation data from abundant real-world footage. To support this goal, we introduce a lightweight data collection pipeline and mobile app that support first-person recording, third-person recording, and synchronized capture of both views. This setup combines accessible single-view collection with paired RGB and camera geometry for learning cross-view correspondence. Building on a pretrained video diffusion transformer, our model incorporates geometric guidance and auxiliary supervision from semantic masks and hand keypoints to facilitate optimization and improve the fidelity of hand-object interactions. Camera geometry is injected stochastically during training and gradually removed, allowing inference without target-camera calibration. To accommodate the diversity of Internet videos, we combine source-video perception and reasoning, optional actor references, and multi-shot adaptation to identify the intended actor and follow an operation across camera cuts. The collection system and generation model provide a foundation for expanding first-person operation data across tasks, scenes, and participants.

The paper's own sentences, arranged, kept beside the review and separate from it. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. No sentence in this record was written by a model.

abstract compiled · 2026-11-13

The authors' main claim, as the classifier picked it out 0.75

“The collection system and generation model provide a foundation for expanding first-person operation data across tasks, scenes, and participants.”

How the main claim is stated: asserted 0.95

What the abstract itself states, as probabilities: claims first 0.05 · states limitation 0.20 · human participants 0.67 · controlled comparison 0.03 · randomized assignment 0.02 · states quantitative result 0.03 · compares with prior results 0.05 · mentions code or data availability 0.06

The abstract, sentence by sentence

This table labels the abstract only (arXiv listing abstract v1). The number is the probability the classifier gave the label.

SentenceRoleThe authors' own workStates the main conclusionStrength
We present VBot Exo2Ego Policy, a video generation model that turns third-person videos into first-person observations of a selected actor, offering a practical way to expand first-person operation data from abundant real-world footage. contribution 1.00 0.92 0.43 asserted 0.99
To support this goal, we introduce a lightweight data collection pipeline and mobile app that support first-person recording, third-person recording, and synchronized capture of both views. method 0.70 0.94 0.04 asserted 1.00
This setup combines accessible single-view collection with paired RGB and camera geometry for learning cross-view correspondence. method 0.95 0.77 0.04 asserted 1.00
Building on a pretrained video diffusion transformer, our model incorporates geometric guidance and auxiliary supervision from semantic masks and hand keypoints to facilitate optimization and improve the fidelity of hand-object interactions. method 0.99 0.89 0.06 asserted 0.99
Camera geometry is injected stochastically during training and gradually removed, allowing inference without target-camera calibration. method 0.99 0.91 0.05 asserted 1.00
To accommodate the diversity of Internet videos, we combine source-video perception and reasoning, optional actor references, and multi-shot adaptation to identify the intended actor and follow an operation across camera cuts. method 0.94 0.87 0.06 asserted 1.00
The collection system and generation model provide a foundation for expanding first-person operation data across tasks, scenes, and participants. interpretation 0.71 0.70 0.46 asserted 0.95

What kind of paper the classifier reports this is: method or tool 1.00 · evidence from observation or experiment 0.81

What was examined, and what was not

Sources read

  • arXiv listing abstract v1

Not examined

  • full text
  • figures and tables
  • references and citing sentences
  • relations to other papers

Levels run

  • P abstract

Not run

  • S sections
  • T body sentences
  • C citations
  • R cross-paper relations

How the labels were made

Compiled by classifier-1.13.0 via TypeSafe API, question bank abstract-p1-20260923, 2026-11-13: 1 request, 5,602 input tokens.

Every quoted sentence on this record is the authors'. Every label is from Pith's closed vocabulary, with the probability shown beside it. This record as JSON

Figures (11)
Figure 1 of the paper
Figure 1. From observed operations to first-person video. Selected EXO inputs and generated EGO frames. 1 arXiv:2611.16417v1 [cs.CV] 12 Nov 2026
Figure 2 of the paper
Figure 2. VBot Exo2Ego Policy architecture. Solid paths generate EGO RGB; dashed paths introduce geometric, semantic, and hand supervision during training. 4.2 Stochastic Camera Conditioning To incorporate calibrated camera geometry into viewpoint learning while preserving the pretrained generator’s behavior at initialization, we introduce stochastic camera conditioning. We sample a Bernoulli gate for each training clip to control whether calibrated target-camera rays are injected through zero-initialized projections. We anneal the probability of ray injection to zero before the end of…
Figure 3 of the paper
Figure 3. Semantic masks and hand readouts from generated EGO. Raiders of the Lost Ark EXO input Generated EGO frame 55 frame 125 The Gold Rush (Chaplin) EXO input Generated EGO (grayscale) frame 55 frame 85
Figure 4 of the paper
Figure 4. Internet EXO-to-EGO examples
Figure 5 of the paper
Figure 5. Training loss and ray availability. The curves average consecutive 100-update blocks; conditioning probability follows the original schedule during ray-free continuation. B Data Alignment and Packaging Paired capture timestamps are mapped to a common time base. The nearest EGO frame to each EXO reference is selected within the recorded 20 ms acceptance window. Camera calibration and image transforms remain aligned to the decoded frames. Dataset records bind source clips, view provenance, text, optional camera rays, and auxiliary targets to their corresponding temporal windows. The…
Figure 6 of the paper
Figure 6. Joint RGB/mask generation on Echo self-collected data
Figure 7 of the paper
Figure 7. Joint RGB/mask generation for bicycle repair
Figure 8 of the paper
Figure 8. Joint RGB/mask generation for basketball
Figure 9 of the paper
Figure 9. Hand observations on HOT3D
Figure 10 of the paper
Figure 10. Hand observations on ARCTIC
Figure 11 of the paper
Figure 11. EXO-to-EGO generation: Star Wars: A New Hope

Figures are the authors', reproduced from the paper.

References (31)
  1. [1] Ben Mildenhall et al. NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis. ECCV, 2020. https://arxiv.org/abs/2003.08934 on PithPith's reviewarXiv
  2. [2] Peng Wang et al. NeuS: Learning Neural Implicit Surfaces by Volume Rendering for Multi-view Reconstruction. NeurIPS, 2021.https://arxiv.org/abs/2106.10689 on PithPith's reviewarXiv
  3. [3] Thomas Müller, Alex Evans, Christoph Schied, and Alexander Keller. Instant Neural Graphics Primitives with a Multiresolution Hash Encoding. ACM Transactions on Graphics, 2022. https://arxiv.org/abs/2201.05989 on PithPith's reviewarXiv
  4. [4] Bernhard Kerbl et al. 3D Gaussian Splatting for Real-Time Radiance Field Rendering. ACM Transactions on Graphics, 2023.https://arxiv.org/abs/2308.04079 on PithPith's reviewarXiv
  5. [5] Lars Mescheder et al. Sharp Monocular View Synthesis in Less Than a Second. arXiv:2512.10685, 2025. https://arxiv.org/abs/2512.10685 on PitharXiv
  6. [6] Lihan Jiang et al. AnySplat: Feed-forward 3D Gaussian Splatting from Unconstrained Views. ACM Transactions on Graphics, 2025.https://arxiv.org/abs/2505.23716 on PitharXiv
  7. [7] Guanjun Wu et al. 4D Gaussian Splatting for Real-Time Dynamic Scene Rendering. CVPR, 2024. https://arxiv.org/abs/2310.08528 on PithPith's reviewarXiv
  8. [8] Shuzhe Wang et al. DUSt3R: Geometric 3D Vision Made Easy. CVPR, 2024. https://arxiv.org/abs/2312.14132 on PithPith's reviewarXiv
  9. [9] Jianyuan Wang et al. VGGT: Visual Geometry Grounded Transformer. CVPR, 2025. https://arxiv.org/abs/2503.11651 on PithPith's reviewarXiv
  10. [10] Haotong Lin et al. Depth Anything 3: Recovering the Visual Space from Any Views. arXiv:2511.10647, 2025. https://arxiv.org/abs/2511.10647 on PithPith's reviewarXiv
  11. [11] Yifan Wang et al.𝜋3: Permutation-Equivariant Visual Geometry Learning. ICLR, 2026. https://arxiv.org/abs/2507.13347 on PithPith's reviewarXiv
  12. [12] Chuhan Zhang et al. Efficiently Reconstructing Dynamic Scenes One D4RT at a Time. arXiv:2512.08924, 2025. https://arxiv.org/abs/2512.08924 on PitharXiv
  13. [13] Yihang Luo et al. 4RC: 4D Reconstruction via Conditional Querying Anytime and Anywhere. arXiv:2602.10094, 2026.https://arxiv.org/abs/2602.10094. VBot Exo2Ego Policy Technical Report 18 / 27 on PithPith's reviewarXiv
  14. [14] Jay Karhade et al. Any4D: Unified Feed-Forward Metric 4D Reconstruction. arXiv:2512.10935, 2025. https://arxiv.org/abs/2512.10935 on PitharXiv
  15. [15] Mark Yu et al. TrajectoryCrafter: Redirecting Camera Trajectory for Monocular Videos via Diffusion Models. ICCV, 2025.https://arxiv.org/abs/2503.05638 on PithPith's reviewarXiv
  16. [16] Xuanchi Ren et al. GEN3C: 3D-Informed World-Consistent Video Generation with Precise Camera Control. arXiv:2503.03751, 2025.https://arxiv.org/abs/2503.03751 on PithPith's reviewarXiv
  17. [17] Tao Hu et al. EX-4D: EXtreme Viewpoint 4D Video Synthesis via Depth Watertight Mesh. arXiv:2506.05554, 2025.https://arxiv.org/abs/2506.05554 on PithPith's reviewarXiv
  18. [18] Jianhong Bai et al. ReCamMaster: Camera-Controlled Generative Rendering from a Single Video. ICCV, 2025. https://arxiv.org/abs/2503.11647 on PithPith's reviewarXiv
  19. [19] Kuan Heng Lin et al. Vista4D: Video Reshooting with 4D Point Clouds. CVPR, 2026. https://arxiv.org/abs/2604.21915 on PithPith's reviewarXiv
  20. [20] Yongqi Mao et al. Manifold4D: Denoising on Point Cloud Rendered Manifolds for Video Re-shooting. arXiv:2608.28174, 2026.https://arxiv.org/abs/2608.28174 on PithPith's reviewarXiv
show 11 more
  1. [21] Taewoong Kang et al. EgoX: Egocentric Video Generation from a Single Exocentric Video. arXiv:2512.08269, 2025.https://arxiv.org/abs/2512.08269 on PitharXiv
  2. [22] Shengze Wang et al. Grounded-Exo2Ego: Structured Semantic Grounding for Robust Exocentric-to-Egocentric Video Generation. arXiv:2608.20534, 2026.https://arxiv.org/abs/2608.20534 on PithPith's reviewarXiv
  3. [23] Kristen Grauman et al. Ego-Exo4D: Understanding Skilled Human Activity from First- and Third-Person Perspectives. CVPR, 2024.https://ego-exo4d-data.org/ on Pith
  4. [24] Yufei Liu et al. ACE-Ego-Hand: Repurposing Video Diffusion Models for Occlusion-Robust Egocentric 3D Hand Motion Recovery. arXiv:2608.20308, 2026.https://arxiv.org/abs/2608.20308 on PithPith's reviewarXiv
  5. [25] Prithviraj Banerjee et al. HOT3D: Hand and Object Tracking in 3D from Egocentric Multi-View Videos. CVPR, 2025.https://arxiv.org/abs/2411.19167 on PithPith's reviewarXiv
  6. [26] Zicong Fan et al. ARCTIC: A Dataset for Dexterous Bimanual Hand-Object Manipulation. CVPR, 2023. https://arxiv.org/abs/2204.13662 on PithPith's reviewarXiv
  7. [27] Gyeongsik Moon et al. A Dataset of Relighted 3D Interacting Hands. NeurIPS Datasets and Benchmarks, 2023. https://arxiv.org/abs/2310.17768 on PithPith's reviewarXiv
  8. [28] Vincent Leroy et al. Grounding Image Matching in 3D with MASt3R. ECCV, 2024. https://arxiv.org/abs/2406.09756 on PithPith's reviewarXiv
  9. [29] Philipp Lindenberger et al. LightGlue: Local Feature Matching at Light Speed. ICCV, 2023. https://arxiv.org/abs/2306.13643 on PithPith's reviewarXiv
  10. [30] Jiaming Sun et al. LoFTR: Detector-Free Local Feature Matching with Transformers. CVPR, 2021. https://arxiv.org/abs/2104.00680 on PithPith's reviewarXiv
  11. [31] Johan Edstedt et al. RoMa: Robust Dense Feature Matching. CVPR, 2024. https://arxiv.org/abs/2305.15404. VBot Exo2Ego Policy Technical Report 19 / 27 A Training and Evaluation Details A.1 Shared Flow Convention The effective velocity uses the base model’s inverse-CFG convention... on PithPith's reviewarXiv

Challenges and discussion

Disagree with the review, or know something it missed? Leave a comment, or challenge this review formally: a challenge becomes a permanent entry in this paper's record, and its decision is published here.

No comments yet.

Sign in to comment