Video Frame Interpolation Transformer
Preprint
Abstract
Existing methods for video interpolation heavily rely on deep convolution neural networks, and thus suffer from their intrinsic limitations, such as content-agnostic kernel weights and restricted receptive field. To address these issues, we propose a Transformer-based video interpolation framework that allows content-aware aggregation weights and considers long-range dependencies with the self-attention operations. To avoid the high computational cost of global self-attention, we introduce the concept of local attention into video interpolation and extend it to the spatial-temporal domain. Furthermore, we propose a space-time separation strategy to save memory usage, which also improves performance. In addition, we develop a multi-scale frame synthesis scheme to fully realize the potential of Transformers. Extensive experiments demonstrate the proposed model performs favorably against the state-of-the-art methods both quantitatively and qualitatively on a variety of benchmark datasets.Keywords
All Related Versions
This publication has 8 references indexed in Scilit:
- FLAVR: Flow-Agnostic Video Representations for Fast Frame Interpolation2023
- SwinIR: Image Restoration Using Swin Transformer2021
- Video Frame Interpolation via Generalized Deformable ConvolutionIEEE Transactions on Multimedia, 2021
- Learning Spatial and Spatio-Temporal Pixel Aggregations for Image and Video DenoisingIEEE Transactions on Image Processing, 2020
- Video Enhancement with Task-Oriented FlowInternational Journal of Computer Vision, 2019
- Motion Blur Kernel Estimation via Deep LearningIEEE Transactions on Image Processing, 2017
- Three-Dimensional Reconstruction of the Digestive Wall in Capsule Endoscopy Videos Using Elastic Video InterpolationIEEE Transactions on Medical Imaging, 2010
- Image quality assessment: from error visibility to structural similarityIEEE Transactions on Image Processing, 2004