Research map / Video / Video segmentation and tracking
Single-Frame Regularization for Temporally Stable CNNs
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 transform-invariance regularization at α=0.95 | Colorization CNN fine-tuning · YouTube-8M colorization | Smoothness: 0.7243 -> 2.8934 | Helped |
| Add transform-invariance regularization at α=0.95 | Colorization CNN fine-tuning · YouTube-8M colorization | PSNR: 18.5805 -> 18.8880 | Helped |
| Add sparse-Jacobian regularization at α=0.95 | Colorization CNN fine-tuning · YouTube-8M colorization | Smoothness: 0.7243 -> 2.5079 | Helped |
| Use α=0.8 rather than α=0.95 for transform invariance (ablation) | Colorization CNN fine-tuning · YouTube-8M colorization | Smoothness: 2.8934 -> 1.9074 | Hurt |
| Use α=0.8 rather than α=0.95 for transform invariance (ablation) | Colorization CNN fine-tuning · YouTube-8M colorization | PSNR: 18.8880 -> 18.9437 | Helped |
| Add transform-invariance regularization at α=0.8 | HDR reconstruction CNN fine-tuning · HDR video reconstruction | Smoothness: 5.9951 -> 8.0798 | Helped |
| Add transform-invariance regularization at α=0.8 | HDR reconstruction CNN fine-tuning · HDR video reconstruction | PSNR: 25.5131 -> 25.4374 | Hurt |
| Add sparse-Jacobian regularization at α=0.95 | HDR reconstruction CNN fine-tuning · HDR video reconstruction | Smoothness: 5.9951 -> 7.3048 | Helped |
| Use α=0.8 rather than α=0.95 for transform invariance (ablation) | HDR reconstruction CNN fine-tuning · HDR video reconstruction | PSNR: 24.1678 -> 25.4374 | Helped |
| Use α=0.8 rather than α=0.95 for transform invariance (ablation) | HDR reconstruction CNN fine-tuning · HDR video reconstruction | Smoothness: 10.6435 -> 8.0798 | Hurt |
| Use noise perturbations instead of geometric transformations for stability regularization (ablation) | CNN temporal regularization · Colorization and HDR reconstruction | Smoothness | Hurt |
| Use naive geometric augmentation instead of explicit temporal regularization (ablation) | CNN training · Colorization and HDR reconstruction | Smoothness | Hurt |
Closest papers by meaning
- Instant Video Models: Universal Adapters for Stabilizing Image-Based Networks (NeurIPS 2025 · similarity 0.89)
- Blind Video Temporal Consistency via Deep Video Prior (NeurIPS 2020 · similarity 0.88)
- Learning Temporal Consistency for Low Light Video Enhancement From Single Images (CVPR 2021 · similarity 0.87)
- Slow and Steady Feature Analysis: Higher Order Temporal Coherence in Video (CVPR 2016 · similarity 0.87)
- Smooth Regularization for Efficient Video Recognition (NeurIPS 2025 · similarity 0.87)
- MEET: Towards Memory-Efficient Temporal Sparse Deep Neural Networks (CVPR 2025 · similarity 0.85)
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.