Video Frame Interpolation Transformer

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.

References in corpus (8)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• UCF101: A Dataset of 101 Human Actions Classes From Videos in The Wild
• Swin Transformer: Hierarchical Vision Transformer using Shifted Windows
• Video Super-Resolution Transformer
• SwinIR: Image Restoration Using Swin Transformer
• Learning Spatial and Spatio-Temporal Pixel Aggregations for Image and Video Denoising
• Quadratic video interpolation
• FLAVR: Flow-Agnostic Video Representations for Fast Frame Interpolation