Unsupervised Geometry Aware Representation For 3d Human Pose Estimation

By switzerlandersing On Sep 11, 2025

GitHub - Hrhodin/UnsupervisedGeometryAwareRepresentationLearning: Unsupervised Geometry-Aware ...

GitHub - Hrhodin/UnsupervisedGeometryAwareRepresentationLearning: Unsupervised Geometry-Aware ... We have introduced an approach to learning a geometry aware representation of the human body in an unsupervised manner, given only multi view imagery. our experiments have shown that this representation is effective both as an intermediate one for 3d pose estimation and for novel view synthesis. Modern 3d human pose estimation techniques rely on deep networks, which require large amounts of training data. in this work, we propose to overcome this problem by learning a geometry aware body representation from multi view images without 3d annotations.

(PDF) Unsupervised Geometry-Aware Representation For 3D Human Pose Estimation

(PDF) Unsupervised Geometry-Aware Representation For 3D Human Pose Estimation We have introduced an approach to learning a geometry aware representation of the human body in an unsupervised manner, given only multi view imagery. our experiments have shown that this representation is effective both as an intermediate one for 3d pose estimation and for novel view synthesis. In this paper, we propose to overcome this problem by learning a geometry aware body representation from multi view images without annotations. to this end, we use an encoder decoder that predicts an image from one viewpoint given an image from another viewpoint. We present an unsupervised learning approach to recover 3d human pose from 2d skeletal joints extracted from a single image. our method does not require any multi view image data, 3d skeletons, correspondences between 2d 3d points, or use previously learned 3d priors during training. In this paper, we propose to overcome this problem by learning a geometry aware body representation from multi view images without annotations. to this end, we use an encoder decoder that predicts an image from one viewpoint given an image from another viewpoint.

Figure 1 From OCR-Pose: Occlusion-aware Contrastive Representation For Unsupervised 3D Human ...

Figure 1 From OCR-Pose: Occlusion-aware Contrastive Representation For Unsupervised 3D Human ... We present an unsupervised learning approach to recover 3d human pose from 2d skeletal joints extracted from a single image. our method does not require any multi view image data, 3d skeletons, correspondences between 2d 3d points, or use previously learned 3d priors during training. In this paper, we propose to overcome this problem by learning a geometry aware body representation from multi view images without annotations. to this end, we use an encoder decoder that predicts an image from one viewpoint given an image from another viewpoint. We present an unsupervised learning approach to re cover 3d human pose from 2d skeletal joints extracted from a single image. our method does not require any multi view image data, 3d skeletons, correspondences between 2d 3d points, or use previously learned 3d priors during training. We will show that our geometry aware latent representation learned from multi view imagery but without annotations allows us to train a 3d pose estimation network using much less labeled data. In this work, we propose a geometry aware 3d repre sentation for the human pose to address this limitation by using multiple views in a simple auto encoder model at the training stage and only 2d keypoint information as super vision.

Unsupervised Geometry-Aware Representation For 3D Human Pose Estimation | DeepAI

Unsupervised Geometry-Aware Representation For 3D Human Pose Estimation | DeepAI We present an unsupervised learning approach to re cover 3d human pose from 2d skeletal joints extracted from a single image. our method does not require any multi view image data, 3d skeletons, correspondences between 2d 3d points, or use previously learned 3d priors during training. We will show that our geometry aware latent representation learned from multi view imagery but without annotations allows us to train a 3d pose estimation network using much less labeled data. In this work, we propose a geometry aware 3d repre sentation for the human pose to address this limitation by using multiple views in a simple auto encoder model at the training stage and only 2d keypoint information as super vision.

ECCV 2018 Unsupervised Geometry-Aware Representation Learning For 3D Human Pose Estimation - YouTube

ECCV 2018 Unsupervised Geometry-Aware Representation Learning For 3D Human Pose Estimation - YouTube In this work, we propose a geometry aware 3d repre sentation for the human pose to address this limitation by using multiple views in a simple auto encoder model at the training stage and only 2d keypoint information as super vision.