English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

3D hand shape and pose estimation from a single depth map is a new andchallenging computer vision problem with many applications. Thestate-of-the-art methods directly regress 3D hand meshes from 2D depth imagesvia 2D convolutional neural networks, which leads to artefacts in theestimations due to perspective distortions in the images. In contrast, wepropose a novel architecture with 3D convolutions trained in aweakly-supervised manner. The input to our method is a 3D voxelized depth map,and we rely on two hand shape representations. The first one is the 3Dvoxelized grid of the shape which is accurate but does not preserve the meshtopology and the number of mesh vertices. The second representation is the 3Dhand surface which is less accurate but does not suffer from the limitations ofthe first representation. We combine the advantages of these tworepresentations by registering the hand surface to the voxelized hand shape. Inthe extensive experiments, the proposed approach improves over the state of theart by 47.8% on the SynHand5M dataset. Moreover, our augmentation policy forvoxelized depth maps further enhances the accuracy of 3D hand pose estimationon real data. Our method produces visually more reasonable and realistic handshapes on NYU and BigHand2.2M datasets compared to the existing approaches.

HandVoxNet: Deep Voxel-Based Network for 3D Hand Shape and Pose Estimation from a Single Depth Map