Research map / Video / Video segmentation and tracking
Towards Explainable Video Camouflaged Object Detection: SAM2 with Eventstream-Inspired Data
Hong Zhang, Yixuan Lyu, Hanyang Liu, Jianbo Song, Ding Yuan, Yifan Yang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add the Prompt Embedding Generator instead of a blank mask prompt (ablation) | SAM2 for VCOD · MoCA-Mask | S-alpha: 0.630 -> 0.640 | Improved |
| add the Prompt Embedding Generator instead of a blank mask prompt (ablation) | SAM2 for VCOD · MoCA-Mask | weighted F-measure: 0.328 -> 0.336 | Improved |
| add eventstream data and Time Bridge Attention on top of prompt generator (ablation) | EventVCOD · MoCA-Mask | weighted F-measure: 0.336 -> 0.573 | Improved |
| full model versus baseline SAM2 (ablation) | EventVCOD · MoCA-Mask | mIoU: 0.303 -> 0.496 | Improved |
| use ESIM synthesized events instead of eventstream-inspired data (ablation) | EventVCOD · MoCA-Mask | S-alpha: .753 -> .722 | Hurt |
| use ESIM synthesized events instead of eventstream-inspired data (ablation) | EventVCOD · MoCA-Mask | acquisition FPS: 35.65 -> 5.73 | Hurt |
| mix COD10K and MoCA-Mask in one training stage instead of two-stage fine-tuning (ablation) | EventVCOD · MoCA-Mask | overall performance | Hurt |
| add a second initial prompt frame (frame 100) for a hard hedgehog video | EventVCOD · failure case (single video) | qualitative localization | Improved |
Closest papers by meaning
- Endow SAM with Keen Eyes: Temporal-spatial Prompt Learning for Video Camouflaged Object Detection (CVPR 2024 · similarity 0.89)
- CamSAM2: Segment Anything Accurately in Camouflaged Videos (NeurIPS 2025 · similarity 0.86)
- LEOD: Label-Efficient Object Detection for Event Cameras (CVPR 2024 · similarity 0.84)
- FlexEvent: Towards Flexible Event-Frame Object Detection at Varying Operational Frequencies (NeurIPS 2025 · similarity 0.84)
- CamoDiffusion: Camouflaged Object Detection via Conditional Diffusion Models (AAAI 2024 · similarity 0.84)
- Event-assisted Low-Light Video Object Segmentation (CVPR 2024 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
