HumanEva: Synchronized Video and Motion Capture Dataset for Evaluation of Articulated Human Motion

By Leonid Sigal, Michael J. Black. 2006

453 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/22d70d143135af112b7852416e416bbd02d093db

How researchers see this paper

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

Citations in the corpus

MuHAVi: A Multicamera Human Action Video Dataset for the Evaluation of Action Recognition Methods (2010, methodology): “The public multi-camera, multi-action dataset is an improvement over existing datasets (e.g. PETS, CAVIAR, soccerdataset) that have not been developed specifically for human action recognition and complements other action recognition datasets (KTH, Weizmann, IXMAS, HumanEva, CMU Motion).”
LabelMe: Online Image Annotation and Applications (2010, background): “containing many hours of television or surveillance data [63], [2], [3], [17], [ 59 ].”
Image Clustering Using Local Discriminant Models and Global Integration (2010, background): “, YALE-B [10], UMIST [14], MSRA [15], JAFFE [21] and Pointing04 [13], one shape image database MPEG7 [3] and one 3D human pose database HumanEva [31].”
Advances in View-Invariant Human Motion Analysis: A Review (2010, background): “), for more details please refer to [101].”
Free-Viewpoint Video of Human Actors Using Multiple Handheld Kinects (2013, background): “The last are markerless motion capture approaches that concentrate on capturing the skeleton motion of the human performance, e.g., [27], [28], [29], [30], [31].”
Factorized Latent Spaces with Structured Sparsity (2010, methodology): “Here, we used the HumanEva dataset [22] which consists of synchronized images and motion capture data describing the 3D locations of the 19 joints of a human skeleton.”
A survey of vision-based methods for action representation, segmentation and recognition (2011, background): “[118] were primarily designed for motion capture and hence only contain very simple action.”
Motion capture using joint skeleton tracking and surface estimation (2009, background): “• Since skeleton-based pose estimation is more constrained than surface estimation, our approach is less sensitive to silhouette noise than comparable visual hull approaches and runs even on medium quality multi-view sequences like the HumanEva benchmark [23].”
Monocular 3D pose estimation and tracking by detection (2010, methodology): “In order to quantitatively evaluate the performance of our 3D pose estimation method we use the “HumanEva II” dataset [23], which provides synchronized images and motion capture data and is a standard evaluation benchmark for 2D and 3D human pose estimation.”
A Review on Human Activity Recognition Using Vision-Based Method (2017, methodology): “This method has been shown to improve performance on the public available dataset HumanEva [109].”
Fusing 2D Uncertainty and 3D Cues for Monocular Body Pose Estimation (2016, background): “6m and fine-tuned on HumanEva.”
Poselets: Body part detectors trained using 3D human pose annotations (2009, result): “It is interesting that we outperform [5] on H3D but get comparable performance on VOC2007.”
Shape-based pedestrian parsing (2011, methodology): “For training exemplars, we use images taken from the HumanEva dataset [27].”
Fast directional chamfer matching (2010, methodology): “The experiments were also performed on the HumanEva dataset [17] which contains video sequences of multiple human subjects performing various activities captured from different viewing directions.”