Unsupervised Video Interpolation Using Cycle Consistency
This addresses the scarcity of high frame rate training videos for video interpolation, enabling adaptation to new domains without additional data, though it builds incrementally on existing unsupervised and pre-trained methods.
The paper tackles the problem of synthesizing high frame rate videos from low frame rate ones without requiring high frame rate training data, using unsupervised techniques based on cycle consistency and pseudo-supervised loss. The result is a significant improvement in pre-trained models on new domains, increasing PSNR from 32.84dB to 33.05dB on Slowflow and from 31.82dB to 32.53dB on Sintel.
Learning to synthesize high frame rate videos via interpolation requires large quantities of high frame rate training videos, which, however, are scarce, especially at high resolutions. Here, we propose unsupervised techniques to synthesize high frame rate videos directly from low frame rate videos using cycle consistency. For a triplet of consecutive frames, we optimize models to minimize the discrepancy between the center frame and its cycle reconstruction, obtained by interpolating back from interpolated intermediate frames. This simple unsupervised constraint alone achieves results comparable with supervision using the ground truth intermediate frames. We further introduce a pseudo supervised loss term that enforces the interpolated frames to be consistent with predictions of a pre-trained interpolation model. The pseudo supervised loss term, used together with cycle consistency, can effectively adapt a pre-trained model to a new target domain. With no additional data and in a completely unsupervised fashion, our techniques significantly improve pre-trained models on new target domains, increasing PSNR values from 32.84dB to 33.05dB on the Slowflow and from 31.82dB to 32.53dB on the Sintel evaluation datasets.