Research map / Continual learning / Domain and prompt-based
Continual Segmentation under Joint Nonstationarity
Prashant Pandey, Himanshu Kumar, Devineni Sri Venkatraya Chowdary, Brejesh Lall
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 JASCL to U-Net | U-Net · Med JASCL-Disjoint (session 1) | Dice: 0.076 -> 0.460 | Improved |
| add JASCL to U-Net | U-Net · Med JASCL-Disjoint (session 5) | Dice: 0.042 -> 0.324 | Improved |
| add JASCL to MedFormer | MedFormer · Med JASCL-Mixed (session 4) | Dice: 0.067 -> 0.228 | Improved |
| add JASCL to SwinUNetr | SwinUNetr · Med JASCL-Mixed (session 4) | Dice: 0.082 -> 0.210 | Improved |
| add JASCL to SAM | SAM ViT-B · Natural-JASCL (session 2) | mIoU: 30.81 -> 31.22 | Improved |
| add JASCL with unlabeled data to GAPS | GAPS · Semi-Supervised Natural-JASCL (session 3) | mIoU: 14.45 -> 25.47 | Improved |
| add JASCL to MedFormer with unlabeled data | MedFormer · Med Semi-Supervised-JASCL (session 5) | Dice: 0.040 -> 0.293 | Improved |
| add gradient adaptive stabilization (GAS) (ablation) | JASCL (MedFormer) · Med JASCL-Mixed | Dice | Improved |
| add prototype anchored supervision (PAS) (ablation) | JASCL + GAPS · Semi-Supervised Natural-JASCL | mIoU | Improved |
| lower epsilon from 1e-6 to 1e-9 (ablation) | JASCL (U-Net) · Med JASCL-Disjoint session 1 | Dice: 0.443 -> 0.482 | Mixed |
| raise noise variance from 1 to 10 (ablation) | JASCL (U-Net) · Med JASCL-Disjoint session 1 | Dice: 0.460 -> 0.408 | Hurt |
| increase shots K from 3 to 5 (ablation) | JASCL (U-Net) · Med JASCL-Disjoint session 1 | Dice: 0.414 -> 0.460 | Improved |
Closest papers by meaning
- Preparing the Future for Continual Semantic Segmentation (ICCV 2023 · similarity 0.92)
- A Survey on Continual Semantic Segmentation: Theory, Challenge, Method and Application (TPAMI 2024 · similarity 0.91)
- CoMFormer: Continual Learning in Semantic and Panoptic Segmentation (CVPR 2023 · similarity 0.89)
- Continual Semantic Segmentation via Repulsion-Attraction of Sparse and Disentangled Latent Representations (CVPR 2021 · similarity 0.89)
- Towards Continual Universal Segmentation (CVPR 2025 · similarity 0.88)
- Beyond Background Shift: Rethinking Instance Replay in Continual Semantic Segmentation (CVPR 2025 · 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.
