Research map / Architectures and efficiency / RNNs and normalization
LSM: Learning Subspace Minimization for Low-Level Vision
Chengzhou Tang, Lu Yuan, Ping Tan
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 |
|---|---|---|---|
| Use LSM trained on the other three tasks instead of PWC-Net for zero-shot stereo matching | Zero-shot stereo matching · FlyingThings3D | EPE: 5.29 -> 2.47 | Improved |
| Train LSM on the other three tasks and apply it zero-shot to optical flow | LSM · Sintel | EPE: 4.6 EPE | Improved |
| Train LSM on the other tasks and apply it zero-shot to video segmentation | LSM · DAVIS-2016 | IoU: 0.682 IoU | Mixed |
| Replace explicit minimization with a CNN implicit solver that predicts the solution (ablation) | LSM · Interactive segmentation | IoU | Hurt |
| Replace explicit minimization with a CNN implicit solver that predicts the solution (ablation) | LSM · Video segmentation | IoU | Hurt |
| Replace explicit minimization with a CNN implicit solver that predicts the solution (ablation) | LSM · Stereo matching | EPE | Hurt |
| Replace explicit minimization with a CNN implicit solver that predicts the solution (ablation) | LSM · Optical flow | EPE | Hurt |
| Remove minimization context from subspace generation (ablation) | LSM · Interactive segmentation, video segmentation, stereo matching, and optical flow | Task performance | Hurt |
| Remove subspace projection during minimization (ablation) | LSM · Interactive segmentation, video segmentation, stereo matching, and optical flow | Task performance | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- PWC-Net — PWC-Net: CNNs for Optical Flow Using Pyramid, Warping, and Cost Volume (CVPR 2018) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- LSNet: See Large, Focus Small (CVPR 2025 · similarity 0.86)
- C-LoRA: Continual Low-Rank Adaptation for Pre-trained Visual Models (NeurIPS 2026 · similarity 0.85)
- LERE: Learning-Based Low-Rank Matrix Recovery with Rank Estimation (AAAI 2024 · similarity 0.85)
- Memory-Efficient LLM Training with Online Subspace Descent (NeurIPS 2024 · similarity 0.84)
- StelLA: Subspace Learning in Low-rank Adaptation using Stiefel Manifold (NeurIPS 2025 · similarity 0.84)
- UNEM: UNrolled Generalized EM for Transductive Few-Shot Learning (CVPR 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.
