CCVS: Context-aware Controllable Video Synthesis

TL;DR

This paper presents CCVS, a self-supervised, controllable video synthesis method that enhances realism and spatial resolution by conditioning on contextual and ancillary information, using autoregressive models, adversarial training, and multimodal control mechanisms.

Contribution

It introduces a novel self-supervised framework for controllable video synthesis that incorporates contextual conditioning, multimodal ancillary information, and a learnable optical flow module for improved realism.

Findings

Achieves high-quality video synthesis with strong spatial and temporal consistency.

Demonstrates flexibility in controlling synthesis through multimodal ancillary inputs.

Outperforms existing methods on multiple benchmarks.

Abstract

This presentation introduces a self-supervised learning approach to the synthesis of new video clips from old ones, with several new key elements for improved spatial resolution and realism: It conditions the synthesis process on contextual information for temporal continuity and ancillary information for fine control. The prediction model is doubly autoregressive, in the latent space of an autoencoder for forecasting, and in image space for updating contextual information, which is also used to enforce spatio-temporal consistency through a learnable optical flow module. Adversarial training of the autoencoder in the appearance and temporal domains is used to further improve the realism of its output. A quantizer inserted between the encoder and the transformer in charge of forecasting future frames in latent space (and its inverse inserted between the transformer and the decoder) adds even more flexibility by affording simple mechanisms for handling multimodal ancillary information for controlling the synthesis process (eg, a few sample frames, an audio track, a trajectory in image space) and taking into account the intrinsically uncertain nature of the future by allowing multiple predictions. Experiments with an implementation of the proposed approach give very good qualitative and quantitative results on multiple tasks and standard benchmarks.