Research map / Detection and segmentation / CNN features
A Generative Appearance Model for End-To-End Video Object Segmentation
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 generative appearance module (ablation) | the proposed video object segmentation network · YouTube-VOS validation | G overall (%): 50.0 -> 66.0 | Helped |
| add the mask-propagation module (ablation) | the proposed video object segmentation network · YouTube-VOS validation | G overall (%): 64.0 -> 66.0 | Helped |
| add Gaussian components for hard examples, beyond one base component per class (ablation) | the generative appearance module · YouTube-VOS validation | G overall (%): 64.4 -> 66.0 | Helped |
| update the appearance model in each frame (ablation) | the generative appearance module · YouTube-VOS validation | G overall (%): 64.9 -> 66.0 | Helped |
| feed appearance scores to fusion without applying SoftMax (ablation) | the generative appearance module · YouTube-VOS validation | G overall (%): 55.8 -> 66.0 | Helped |
| backpropagate through appearance-model inference during training (ablation) | the proposed video object segmentation network · YouTube-VOS validation | G overall (%): 58.8 -> 66.0 | Helped |
Closest papers by meaning
- Learning Fast and Robust Target Models for Video Object Segmentation (CVPR 2020 · similarity 0.92)
- Learning Video Object Segmentation From Static Images (CVPR 2017 · similarity 0.88)
- Breaking the "Object" in Video Object Segmentation (CVPR 2023 · similarity 0.88)
- Emergence of Object Segmentation in Perturbed Generative Models (NeurIPS 2019 · similarity 0.87)
- Learning Hierarchical Image Segmentation For Recognition and By Recognition (ICLR 2024 · similarity 0.85)
- LoSh: Long-Short Text Joint Prediction Network for Referring Video Object Segmentation (CVPR 2024 · similarity 0.85)
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.