Research map / Video / Video segmentation and tracking
Intra-Class Consistency Guided Class-Agnostic Event Segmentation
Zhipeng Sui, Haiqing Hao, Weihua He, Wenhui Wang
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 brightness representation (ablation) | IC2-ESeg · RGBE-SEG prompt mode | mIoU: 0.55 -> 0.65 | Improved |
| add motion representation (ablation) | IC2-ESeg · RGBE-SEG prompt mode | mIoU: 0.61 -> 0.65 | Improved |
| add noise representation (ablation) | IC2-ESeg · RGBE-SEG prompt mode | mIoU: 0.64 -> 0.65 | Improved |
| add residual connection in lightweight information injection (ablation) | IC2-ESeg · RGBE-SEG prompt mode | mIoU: 0.62 -> 0.65 | Improved |
| add adapter-based visual prompts (ablation) | IC2-ESeg · RGBE-SEG prompt mode | mIoU: 0.61 -> 0.65 | Improved |
| add feature-level distillation loss L_sim (ablation) | IC2-ESeg · RGBE-SEG prompt mode | mIoU: 0.57 -> 0.65 | Improved |
| add pseudo-label supervision loss L_seg (ablation) | IC2-ESeg · RGBE-SEG prompt mode | mIoU: 0.48 -> 0.65 | Improved |
| use handcrafted ICM instead of CNN module (ablation) | IC2-ESeg · RGBE-SEG prompt mode | mIoU: 0.63 -> 0.65 | Improved |
| use handcrafted ICM instead of Transformer module (ablation) | IC2-ESeg · RGBE-SEG prompt mode | mIoU: 0.60 -> 0.65 | Improved |
| add ICM to event baseline | Event features · RGBE-SEG | mean cosine similarity: 0.77 -> 0.80 | Improved |
| add noise representation under slow motion (ablation) | IC2-ESeg · Slow motion | mIoU: 0.612 -> 0.636 | Improved |
| add noise representation under fast motion (ablation) | IC2-ESeg · Fast motion | mIoU: 0.653 -> 0.658 | No clear effect |
Closest papers by meaning
- Efficient Event-Based Semantic Segmentation via Exploiting Frame-Event Fusion: A Hybrid Neural Network Approach (AAAI 2025 · similarity 0.86)
- ESEG: Event-Based Segmentation Boosted by Explicit Edge-Semantic Guidance (AAAI 2025 · similarity 0.86)
- OpenESS: Event-based Semantic Scene Understanding with Open Vocabularies (CVPR 2024 · similarity 0.85)
- Event-assisted Low-Light Video Object Segmentation (CVPR 2024 · similarity 0.85)
- The Shape of Events: Edge-Based Inductive Biases via Cross-Domain Distillation (NeurIPS 2026 · similarity 0.84)
- Know Where You Are From: Event-Based Segmentation via Spatio-Temporal Propagation (AAAI 2025 · 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.
