26 datasets found

A non-parametric approach for video generation from a single video, outperforming single-video GANs in visual quality and realism.

A high-resolution landscape video dataset with audio-visual pairs for sound-guided video generation task.

The SoloDance dataset contains 179 solo dance videos in real scenes collected online.

The iPER dataset was proposed by [25], which was collected in the laboratory environment.

Human Video Motion Transfer (HVMT) aims to, given an image of a source person, generate his/her video that imitates the motion of the driving person.

E2VID is an event-to-video pipeline that converts event data into a video sequence.

The CLEVRER dataset is a dataset for video reasoning, where videos are presented to the model, along with a set of related questions, and the model's outputs are the answers to...

Sky Time-lapse for video generation

TaiChi-HD for video generation

StyleGAN-V for video generation

DIGAN for video generation

MoCoGAN-HD for video generation

Video Implicit Diffusion Models for video generation

WebVid dataset is used for text-to-video generation tasks.

UCF-101, Sky Time-lapse, and Taichi datasets are used for video generation tasks.

Diffusion models have recently achieved remarkable results for video generation. Despite the encouraging performances, the generated videos are typically constrained to a small...

The DeepHDRVideo dataset encompasses both real-world dynamic scenes and static scenes enhanced with random global motion.

A dataset for video generation and video-based tasks.

Videofusion: Decomposed diffusion models for high-quality video generation.

The Kinetics-400 and Kinetics-600 datasets are video understanding datasets used for learning rich and multi-scale spatiotemporal semantics from high-dimensional videos.