Softmax Splatting for Video Frame Interpolation

TL;DR

This paper introduces softmax splatting, a novel differentiable forward warping technique for video frame interpolation, enabling better handling of pixel conflicts and achieving state-of-the-art results.

Contribution

The paper presents softmax splatting as a new differentiable forward warping method that improves video frame interpolation accuracy and flexibility.

Findings

Achieves state-of-the-art frame interpolation results.

Effectively handles multiple pixels mapping to the same location.

Enables fine-tuning of optical flow and feature pyramids.

Abstract

Differentiable image sampling in the form of backward warping has seen broad adoption in tasks like depth estimation and optical flow prediction. In contrast, how to perform forward warping has seen less attention, partly due to additional challenges such as resolving the conflict of mapping multiple pixels to the same target location in a differentiable way. We propose softmax splatting to address this paradigm shift and show its effectiveness on the application of frame interpolation. Specifically, given two input frames, we forward-warp the frames and their feature pyramid representations based on an optical flow estimate using softmax splatting. In doing so, the softmax splatting seamlessly handles cases where multiple source pixels map to the same target location. We then use a synthesis network to predict the interpolation result from the warped representations. Our softmax splatting allows us to not only interpolate frames at an arbitrary time but also to fine tune the feature pyramid and the optical flow. We show that our synthesis approach, empowered by softmax splatting, achieves new state-of-the-art results for video frame interpolation.