Stochastic Image-to-Video Synthesis using cINNs

By Michael Dorkenwald, Timo Milbich, Andreas Blattmann, Robin Rombach, Konstantinos G. Derpanis, Björn Ommer. 2021

67 citations recorded by the source corpus.

Video understanding calls for a model to learn the characteristic interplay between static scene content and its dynamics: Given an image, the model must be able to predict a future progression of the portrayed scene and, conversely, a video should be explained in terms of its static image content and all the remaining characteristics not present in the initial frame. This naturally suggests a bijective mapping between the video domain and the static content as well as residual information. In contrast to common stochastic image-to-video synthesis, such a model does not merely generate arbitrary videos progressing the initial image. Given this image, it rather provides a one-to-one mapping between the residual vectors and the video with stochastic outcomes when sampling. The approach is naturally implemented using a conditional invertible neural network (cINN) that can explain videos by independently modelling static and other video characteristics, thus laying the basis for controlled video synthesis. Experiments on diverse video datasets demonstrate the effectiveness of our approach in terms of both the quality and diversity of the synthesized results. Our project page is available at https://bit.ly/3dg90fV.

Subfields: Computer Vision and Pattern Recognition, Generative Models

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Ying Shan
2. Luc Van Gool
3. Chen Change Loy
4. Xintao Wang
5. Dacheng Tao
6. Yu Qiao
7. Ming-Hsuan Yang
8. Björn Ommer

In Generative Models

1. Andreas Blattmann
2. Ying Shan
3. Xintao Wang
4. Karsten Kreis
5. Robin Rombach
6. Tim Dockhorn
7. Björn Ommer
8. Yong Zhang

Citations in the corpus

• iPOKE: Poking a Still Image for Controlled Stochastic Video Synthesis (2021, methodology): “To realize the conditionally bijective nature of our mapping τθ, we implement it as a conditional invertible neural network (cINN) [56, 18, 17, 62, 19], which requires equal dimensionality of the transformed random variables.”
• Driving Into the Future: Multiview Visual Forecasting and Planning with World Model for Autonomous Driving (2023, methodology): “Various generation methods have been proposed in the past, including VAE-based (Variational Autoen-coder) [16, 33, 58, 61], GAN-based (generative adversarial networks) [3, 15, 31, 49, 56, 70], flow-based [10, 34] and auto-regressive models [18, 64, 68].”
• Align Your Latents: High-Resolution Video Synthesis with Latent Diffusion Models (2023, background): “15 [13] Michael Dorkenwald, Timo Milbich, Andreas Blattmann, Robin Rombach, Konstantinos G.”
• NExT-GPT: Any-to-Any Multimodal LLM (2023, background): “…tasks, such as Image/Video Captioning [99, 16, 56, 56, 27, 49], Image/Video Question Answering [94, 90, 48, 98, 3], Text-to-Image/Video/Speech Synthesis [74, 30, 84, 23, 17, 51, 33], Image-to-Video Synthesis [18, 37] and more, all of which have experienced rapid advancements in past decades.”
• DynamiCrafter: Animating Open-domain Images with Video Diffusion Priors (2023, methodology): “…methods [9, 37, 51, 55, 63– 65, 79 ] transfer motion or appearance information from Although existing approaches has achieved impressive performance, they primarily focus on animating motions in curated domains, particularly stochastic [5, 10, 14, 16, 36, 40, 51, 87] and oscillating [42] motion.”