14 datasets found

The EgoHumans dataset consists of approximately 1 hour of video of up to 5 subjects performing highly dynamic activities.

The Campus and Shelf datasets contain approximately 4000 frames of a single scene.

The CMU Panoptic Studio dataset is a large multi-view pose estimation dataset with several synchronized camera sequences of multiple interacting subjects.

Face detection, pose estimation, and landmark localization in the wild.

Transparent objects are encountered frequently in our daily lives, yet recognizing them poses challenges for conventional vision sensors due to their unique material properties,...

A novel benchmark for out-of-domain robustness and testing common image corruptions for pose estimation.

HetACUMN consists of two tasks: variational image reconstruction and conditional pose prediction. The variational image reconstruction task follows the standard self-supervised...

A dataset for visual odometry, containing 11 sequences with provided ground-truth poses.

The KITTI odometry dataset is a collection of 22 sequences, containing point clouds, images, and GPS recordings of inner-city traffic, residential areas, highway scenes, and...

The dataset used for training the D2-Net model, consisting of 327,036 image pairs with at least 50% overlap in the sparse SfM point cloud.

A dataset for pose estimation and tracking for ASIST system

DeepLabCut dataset for pose estimation

A face detection benchmark with variations in pose, scale, aspect ratio, occlusion and illumination conditions.

Large scale datasets [18, 17, 27, 6] boosted text conditional image generation quality. However, in some domains it could be difficult to make such datasets and usually it could...