- Rating
- 2
- Confidence
- 3
- Informativeness
- 3
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
PhysAtObj learns scene-specific object dynamics from calibrated multi-view videos using object-labelled Gaussians, motion reconstruction, and a time-independent interaction model. Future-frame results are promising, but the evidence for transferable physical laws remains limited.
Critical strengths
- Object-level modeling and the separation of rigid motion from deformation are well motivated.
- Table 1 shows substantial gains, including Moving-Static mIoU increasing from 40.105 to 67.255, with foreground metrics reducing background dominance.
- The appendix specifies state-aware propagation, and the elastic-cube ablation supports its usefulness for deformation.
Critical weaknesses
- The practical use is unclear given costly multi-view capture, scene-specific fitting, and limited evidence of cross-scene generalization.
- The external-field model lacks position and time conditioning, while a single contact normal per object pair may limit complex interactions.
- The small-scale evaluation, largely conducted in controlled settings with few moving objects, provides limited evidence of robustness in complex scenes.
Questions for authors
- Table 1 reports F1 of 87.229 above the precision/recall mean of 85.423, which is incompatible with standard F1 under identical aggregation.
- Section 4 reports 7 real scenes, whereas Appendix A.7 reports 6. Maybe reconcile the counts and list scene IDs for each aggregate.
3 comments
Response to Reviewer UoAW (Part 2)
Q3: The small-scale evaluation, largely conducted in controlled settings ... robustness in complex scenes.
A3: We thank the reviewer for this comment and have added a dedicated analysis in Appendix A.16 (Table 11 and Figure 14; also see the following Table 8), showing that:
- Across our benchmark datasets, the evaluated scenes feature up to 5 foreground entities, 4 simultaneously moving objects, and 5 distinct multi-body interaction events per scene, encompassing 10 core physical phenomena (e.g., elastic coupling, deformable objects, and oblique projectiles).
- Beyond idealized, noise-free settings, our benchmark explicitly incorporates real-world challenges, including scenes with degraded segmentation masks (boundary imperfections) alongside environmental disturbances such as sensor jitter, lighting variations, and visual clutter. Moreover, Figure 14 presents representative frames and masks from the evaluated PhysInOne scenes, demonstrating our model's robust physical reasoning under severe occlusions and intricate contact dynamics.
Overall, our comprehensive benchmark evaluation clearly demonstrates the robustness and breadth of our experiments.
Table 8: The complexity of all evaluated datasets.
| Dataset Partition | # Scenes | # Foreground Objects | # Moving Objects | # Unique Interactions | Key Characteristics & Phenomena Covered | Noisy Masks & Cameras Settings |
|---|---|---|---|---|---|---|
| Moving-Static Object Interaction Dataset | 4 | Moving–stationary collision Moving–fixed collision Moving–moving collision Vertical fall Friction stop Oblique projectile Balloon tether Floating balloon Elastic coupling Deformable object | 6 Mask noise scenes (boundary degradation, segmentation errors) 6 Lighting & Sensor noise scenes (visual clutter, jitter) | |||
| Moving-Moving Object Interaction Dataset | 6 | |||||
| PhysInOne Dataset | 7 | |||||
| Real-world Object Interaction Dataset | 6 | |||||
| Overall Benchmark Span | 23 | Up to 5 | Up to 4 | Up to 5 | 10 physical phenomena | noisy real world scenes |
Q4: Table 1 reports F1 of 87.229 above the precision/recall mean of 85.423, which is incompatible with standard F1 under identical aggregation.
A4: Thank you for pointing out this discrepancy in Table 1.
Upon careful verification, we confirmed that this was an unintentional transcription error during the manual compilation of Table 1. Specifically, the recall and F1-score for one scene in the Moving-Moving Object Interaction Dataset were misrecorded. After incorporating the corrected values, the recall updates from to , and the F1-score updates from to , restoring full numerical consistency across the table.
Additionally, we have thoroughly audited all other experimental results, tables, and metrics throughout the manuscript to ensure complete accuracy.
We have updated Table 1 in the revised paper.
Q5: Section 4 reports 7 real scenes, whereas Appendix A.7 reports 6. Maybe reconcile the counts and list scene IDs for each aggregate.
A5: We sincerely appreciate your careful reading. Yes, that was a typo; there are 6 real scenes in total, which aligns with Appendix A.7.
We have corrected Section 4 and summarized the scene IDs for each aggregate group in Table 12 of the revised paper.
Waiting for discussion
Dear Reviewers UoAW,
Since the discussion is closing very soon (in less than 9 hours), we are still waiting for your thoughts on our rebuttal and revised manuscript. We would be grateful if you could share your further feedback. Thank you for your time.
Best,
Authors
Response to Reviewer UoAW (Part 1)
Q1: The practical use is unclear given costly multi-view capture, scene-specific fitting, and limited evidence of cross-scene generalization.
A1: We thank the reviewer for this comment. We respectfully highlight that learning physical dynamics is notoriously challenging and remains in its early stages. Consequently, current evaluation settings are necessarily controlled, relying on calibrated cameras and per-scene optimization, akin to the early stages of 3D/4D reconstruction such as NeRF and 3DGS. Despite these constraints, we would emphasize our method's generalization capabilities and discuss practical deployment pathways as follows:
In the revised paper, we have updated the conclusion section to better highlight these practical considerations and future directions.
Q2: The external-field model lacks position ... a single contact normal per object pair may limit complex interactions.
A2: We appreciate this insightful comment and clarify our design principles as follows: