MEEV: Body Mesh Estimation On Egocentric Video
This technical report introduces our solution, MEEV, proposed to the EgoBody Challenge at ECCV 2022. Captured from head-mounted devices, the dataset consists of human body shape and motion of interacting people. The EgoBody dataset has challenges such as occluded body or blurry image. In order to overcome the challenges, MEEV is designed to exploit multiscale features for rich spatial information. Besides, to overcome the limited size of dataset, the model is pre-trained with the dataset aggregated 2D and 3D pose estimation datasets. Achieving 82.30 for MPJPE and 92.93 for MPVPE, MEEV has won the EgoBody Challenge at ECCV 2022, which shows the effectiveness of the proposed method. The code is available at https://github.com/clovaai/meev
Code (1)
Tasks
3D human pose and shape estimation3D Human Pose Estimation3D Pose EstimationPose EstimationSimilar Papers 제목 키워드 기반
Social EgoMesh Estimation
Accurately estimating the 3D pose of the camera wearer in egocentric video sequences is crucial to modeling human behavior in virtual and augmented reality applications. The task presents unique challenges due to the lim…
Pose EstimationEgocentric Whole-Body Human Mesh Recovery with Prior-Guided Learning
Egocentric human mesh recovery (HMR) from monocular head-mounted cameras is increasingly important for AR/VR applications, but remains challenging due to the lack of reliable ground-truth (GT) annotations based on parame…
Human Mesh RecoveryPose EstimationFish2Mesh Transformer: 3D Human Mesh Recovery from Egocentric Vision
Egocentric human body estimation allows for the inference of user body pose and shape from a wearable camera's first-person perspective. Although research has used pose estimation techniques to overcome self-occlusions a…
Human Mesh RecoveryPose EstimationPosition4D Human Body Capture from Egocentric Video via 3D Scene Grounding
We introduce a novel task of reconstructing a time series of second-person 3D human body meshes from monocular egocentric videos. The unique viewpoint and rapid embodied camera motion of egocentric videos raise additiona…
Time SeriesTime Series AnalysisEgo-Body Pose Estimation via Ego-Head Pose Estimation
Estimating 3D human motion from an egocentric video sequence plays a critical role in human behavior understanding and has various applications in VR/AR. However, naively learning a mapping between egocentric videos and …
BenchmarkingDisentanglementHead Pose EstimationPose Estimation