Research map / Continual learning / Class-incremental
Continual Learning for Visual Search With Backward Consistent Feature Embedding
Timmy S. T. Wan, Jun-Cheng Chen, Tzer-Yi Wu, Chu-Song Chen
Check 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 inter-session data coherence to intra-session discrimination (ablation) | CVS · CIFAR100, Stanford Dog, and Product-M general-incremental | AR@1: 1.67% to 11.37% | Improved |
| Add neighbor-session model coherence alone to intra-session discrimination (ablation) | CVS · CIFAR100, Stanford Dog, and Product-M general-incremental | AR@1 | Mixed |
| Add exemplar replay to inter-session data coherence (ablation) | CVS · CIFAR100, Stanford Dog, and Product-M general-incremental | AR@1: 1.08% and 7.66% | Improved |
| Add neighbor-session model coherence to Lᶜ + Lᵈ (ablation) | CVS · CIFAR100 and Stanford Dog general-incremental | AR@1 | Improved |
| Add neighbor-session model coherence to Lᶜ + Lᵈ (ablation) | CVS · Product-M general-incremental | AR@1 | Hurt |
| Use CVS instead of the strongest reported baseline | Visual search continual learning · CIFAR100 disjoint, blurry, and general-incremental | AR@1 | Improved |
| Use CVS instead of the strongest reported baseline | Visual search continual learning · Tiny ImageNet disjoint, blurry, and general-incremental | AR@1 | Improved |
| Use CVS instead of other continual learning methods | Visual search continual learning · Stanford Dog general-incremental | AR@4 | Hurt |
Closest papers by meaning
- Not Just Selection, but Exploration: Online Class-Incremental Continual Learning via Dual View Consistency (CVPR 2022 · similarity 0.88)
- Rethinking Query-based Transformer for Continual Image Segmentation (CVPR 2025 · similarity 0.87)
- Enhancing Visual Continual Learning with Language-Guided Supervision (CVPR 2024 · similarity 0.86)
- Resurrecting Old Classes with New Data for Exemplar-Free Continual Learning (CVPR 2024 · similarity 0.86)
- Model Inversion with Layer-Specific Modeling and Alignment for Data-Free Continual Learning (NeurIPS 2025 · similarity 0.86)
- LoRanPAC: Low-rank Random Features and Pre-trained Models for Bridging Theory and Practice in Continual Learning (ICLR 2025 · similarity 0.86)
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.
