Research map / Label-efficient and robust learning / Self-supervised and contrastive
Learning by Distillation: A Self-Supervised Learning Framework for Optical Flow Estimation
Pengpeng Liu, Michael Rung-Tsong Lyu, Irwin King, Jia Xu
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 occlusion handling (ablation) | DistillFlow · KITTI 2012 train | EPE-all: 7.33 -> 3.22 | Improved |
| add edge-aware smoothness (ablation) | DistillFlow · KITTI 2012 train | EPE-all: 3.22 -> 2.92 | Improved |
| add data distillation (ablation) | DistillFlow · KITTI 2012 train | EPE-all: 2.92 -> 1.46 | Improved |
| add data distillation (ablation) | DistillFlow · KITTI 2015 train | EPE-all: 6.45 -> 3.20 | Improved |
| add model distillation (multi-teacher ensemble) (ablation) | DistillFlow with data distillation · KITTI 2012 train | EPE-all: 1.46 -> 1.38 | Improved |
| add model distillation without data distillation (ablation) | DistillFlow · KITTI 2012 train | EPE-all: 2.92 -> 2.86 | Improved |
| add data distillation (ablation) | DistillFlow · Sintel Final raw | EPE-all: 4.41 -> 3.98 | Improved |
| add knowledge distillation (ablation) | FlowNetS · KITTI 2015 train | EPE-all: 8.85 -> 6.33 | Improved |
| add knowledge distillation (ablation) | FlowNetC · KITTI 2015 train | EPE-all: 8.11 -> 5.47 | Improved |
| add semi-supervised distillation with self-annotated data (ablation) | DistillFlow fine-tuning · KITTI 2015 validation | EPE-all: 1.50 -> 1.44 | Improved |
| use distillation v4 instead of v1 (DDFlow) (ablation) | DistillFlow · KITTI 2015 train | EPE-all: 3.88 -> 2.93 | Improved |
| use census transform instead of pixel brightness photometric loss (ablation) | DistillFlow · KITTI 2012 train | EPE-all: 6.58 -> 3.22 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 on most shared results.
- SelFlow — SelFlow: Self-Supervised Learning of Optical Flow (CVPR 2019) · better on 8 of 8 shared results
- MaskFlowNet — MaskFlownet: Asymmetric Feature Matching With Learnable Occlusion Mask (CVPR 2020) · better on 1 of 3 shared results
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
- DDFlow: Learning Optical Flow with Unlabeled Data Distillation (AAAI 2019 · similarity 0.94)
- SelFlow: Self-Supervised Learning of Optical Flow (CVPR 2019 · similarity 0.91)
- Learning by Analogy: Reliable Supervision From Transformations for Unsupervised Optical Flow Estimation (CVPR 2020 · similarity 0.89)
- ZeroFlow: Scalable Scene Flow via Distillation (ICLR 2024 · similarity 0.89)
- OCAI: Improving Optical Flow Estimation by Occlusion and Consistency Aware Interpolation (CVPR 2024 · similarity 0.89)
- Taming generative video models for zero-shot optical flow extraction (NeurIPS 2025 · similarity 0.88)
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.
