Research map / Video / Video segmentation and tracking
Online Video Object Segmentation via Convolutional Trident Network
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 |
|---|---|---|---|
| Use CTN with three decoders and MRF optimization instead of conventional segmentation algorithms | Online video object segmentation · DAVIS | RS mean: 0.665 -> 0.755 | Helped |
| Use CTN with three decoders and MRF optimization instead of conventional segmentation algorithms | Online video object segmentation · DAVIS | RS recall: 0.778 -> 0.890 | Helped |
| Use sequential online segmentation rather than the strongest listed baseline | Online video object segmentation · DAVIS | RS decay: 0.031 -> 0.144 | Hurt |
| Use CTN with three decoders and MRF optimization instead of conventional segmentation algorithms | Online video object segmentation · DAVIS | CA mean: 0.656 -> 0.714 | Helped |
| Use the fast optical flow technique [26] instead of [19] | Prop-Q · DAVIS | RS mean: 0.755 -> 0.734 | Hurt |
| Add definite foreground and background decoders (ablation) | CTN · DAVIS | RS mean: 0.684 -> 0.755 | Helped |
| Add definite foreground and background decoders (ablation) | CTN · DAVIS | CA mean: 0.677 -> 0.714 | Helped |
| Add the separative decoder to select initial foreground pixels (ablation) | CTN · DAVIS | RS mean: 0.663 -> 0.755 | Helped |
| Add the separative decoder to select initial foreground pixels (ablation) | CTN · DAVIS | CA mean: 0.665 -> 0.714 | Helped |
| Apply MRF optimization to refine the segmentation (ablation) | CTN output · DAVIS | RS mean: 0.715 -> 0.755 | Helped |
| Apply MRF optimization to refine the segmentation (ablation) | CTN output · DAVIS | CA mean: 0.663 -> 0.714 | Helped |
| Use Prop-Q instead of the strongest reported SegTrack baseline | Online video object segmentation · SegTrack | Jaccard index (average): 0.80 -> 0.68 | Hurt |
Closest papers by meaning
- Fast and Accurate Online Video Object Segmentation via Tracking Parts (CVPR 2018 · similarity 0.89)
- Learning Video Object Segmentation From Static Images (CVPR 2017 · similarity 0.88)
- From ViT Features to Training-free Video Object Segmentation via Streaming-data Mixture Models (NeurIPS 2023 · similarity 0.88)
- BubbleNets: Learning to Select the Guidance Frame in Video Object Segmentation by Deep Sorting Frames (CVPR 2019 · similarity 0.87)
- One-Shot Video Object Segmentation (CVPR 2017 · similarity 0.87)
- Fast Video Object Segmentation by Reference-Guided Mask Propagation (CVPR 2018 · similarity 0.87)
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.