Research map / Video / Video segmentation and tracking
Fast Spatio-Temporal Residual Network for Video Super-Resolution
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 factorized fast spatio-temporal residual blocks (FRBs) (ablation) | FSTRN baseline without FRB, CRL, and LRL · Video SR, ×4 | PSNR | Helped |
| Add LR-space residual learning (LRL) (ablation) | FSTRN with FRBs · Video SR, ×4 | PSNR | Helped |
| Add cross-space residual learning (CRL) using bilinear interpolation (ablation) | FSTRN with FRBs and LRL · Video SR, ×4 | PSNR | Helped |
| Replace bilinear CRL interpolation with nearest, bicubic, or area interpolation (ablation) | FSTRN with GRL · Video SR, ×4 | PSNR | No clear effect |
| Use deconvolution-based CRL interpolation instead of fixed interpolation (ablation) | FSTRN with GRL · Video SR, ×4 | PSNR | No clear effect |
| Factorize a single 3D convolution into spatial and temporal convolutions (ablation) | One residual block · 64 input/output feature maps; 5 frames, 32×32 | Parameters: 55.86% | Helped |
| Factorize a single 3D convolution into spatial and temporal convolutions (ablation) | One residual block · 64 input/output feature maps; 5 frames, 32×32 | FLOPs: 55.48% | Helped |
| Use FSTRN instead of VESPCN | Video SR · Dancing, Treadmill, Flag, Fan, Turbine (Average, ×4) | PSNR: +0.55 dB | Helped |
| Use FSTRN instead of VESPCN | Video SR · Dancing, Treadmill, Flag, Fan, Turbine (Average, ×4) | SSIM: 0.2 | Helped |
Closest papers by meaning
- Residual Invertible Spatio-Temporal Network for Video Super-Resolution (AAAI 2019 · similarity 0.93)
- Real-Time Video Super-Resolution With Spatio-Temporal Networks and Motion Compensation (CVPR 2017 · similarity 0.92)
- Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network (CVPR 2016 · similarity 0.89)
- Temporal Modulation Network for Controllable Space-Time Video Super-Resolution (CVPR 2021 · similarity 0.89)
- Building an End-to-End Spatial-Temporal Convolutional Network for Video Super-Resolution (AAAI 2017 · similarity 0.89)
- SAVSR: Arbitrary-Scale Video Super-Resolution via a Learned Scale-Adaptive Network (AAAI 2024 · 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.