Mitigating Surgical Data Imbalance with Dual-Prediction Video Diffusion Model
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866909832196390912 |
|---|---|
| author | Venkatesh, Danush Kumar Schmidt, Adam Jamal, Muhammad Abdullah Mohareri, Omid |
| author_facet | Venkatesh, Danush Kumar Schmidt, Adam Jamal, Muhammad Abdullah Mohareri, Omid |
| contents | Surgical video datasets are essential for scene understanding, enabling procedural modeling and intra-operative support. However, these datasets are often heavily imbalanced, with rare actions and tools under-represented, which limits the robustness of downstream models. We address this challenge with $SurgiFlowVid$, a sparse and controllable video diffusion framework for generating surgical videos of under-represented classes. Our approach introduces a dual-prediction diffusion module that jointly denoises RGB frames and optical flow, providing temporal inductive biases to improve motion modeling from limited samples. In addition, a sparse visual encoder conditions the generation process on lightweight signals (e.g., sparse segmentation masks or RGB frames), enabling controllability without dense annotations. We validate our approach on three surgical datasets across tasks including action recognition, tool presence detection, and laparoscope motion prediction. Synthetic data generated by our method yields consistent gains of 10-20% over competitive baselines, establishing $SurgiFlowVid$ as a promising strategy to mitigate data imbalance and advance surgical video understanding methods. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_07345 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Mitigating Surgical Data Imbalance with Dual-Prediction Video Diffusion Model Venkatesh, Danush Kumar Schmidt, Adam Jamal, Muhammad Abdullah Mohareri, Omid Quantitative Methods Artificial Intelligence Image and Video Processing Surgical video datasets are essential for scene understanding, enabling procedural modeling and intra-operative support. However, these datasets are often heavily imbalanced, with rare actions and tools under-represented, which limits the robustness of downstream models. We address this challenge with $SurgiFlowVid$, a sparse and controllable video diffusion framework for generating surgical videos of under-represented classes. Our approach introduces a dual-prediction diffusion module that jointly denoises RGB frames and optical flow, providing temporal inductive biases to improve motion modeling from limited samples. In addition, a sparse visual encoder conditions the generation process on lightweight signals (e.g., sparse segmentation masks or RGB frames), enabling controllability without dense annotations. We validate our approach on three surgical datasets across tasks including action recognition, tool presence detection, and laparoscope motion prediction. Synthetic data generated by our method yields consistent gains of 10-20% over competitive baselines, establishing $SurgiFlowVid$ as a promising strategy to mitigate data imbalance and advance surgical video understanding methods. |
| title | Mitigating Surgical Data Imbalance with Dual-Prediction Video Diffusion Model |
| topic | Quantitative Methods Artificial Intelligence Image and Video Processing |
| url | https://arxiv.org/abs/2510.07345 |