Efficient and precise interactive hand tracking through joint, continuous optimization of pose and correspondences

By Jonathan Taylor, Lucas Bordeaux, Thomas Joseph Cashman, Bob Corish, Cem Keskin, Toby Sharp, Eduardo Soto, David Sweeney, Julien P. C. Valentin, B. Luff and 8 more. ACM Transactions on Graphics, 2016

314 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Control and Systems Engineering, Human-Computer Interaction

View Paper: https://www.semanticscholar.org/paper/b64cf0ff488e6b770ae40b62d5df1c76d7fb8ed5

How researchers see this paper

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

In Computer Vision and Pattern Recognition

In Control and Systems Engineering

In Human-Computer Interaction

Citations in the corpus

Weakly-Supervised 3D Hand Pose Estimation from Monocular RGB Images (2018, background): “In general, they can be classified into generative approaches [20, 34, 16], discriminative approaches [13, 40, 6, 7, 5, 8] and hybrid approaches [25, 31, 30].”
EM-Fusion: Dynamic Object-Level SLAM With Probabilistic Data Association (2019, background): “Dynamic SLAM: Research on tracking and reconstruction of articulated objects such as human body parts [23, 24] or robots [18, 6] is related to dynamic SLAM.”
GANerated Hands for Real-Time 3D Hand Tracking from Monocular RGB (2017, background): “Hybrid methods that combine the best of both generative and discriminative methods show the best performance on benchmark datasets [47, 39, 40, 20, 59].”
Sim2real transfer learning for 3D human pose estimation: motion to the rescue (2019, background): “Again, depth has proven useful [49, 76, 80], where state-of-the-art algorithms typically incorporate some form of generative model in-the-loop at test time.”
Co-fusion: Real-time segmentation, tracking and fusion of multiple objects (2017, background): “These include general articulated tracking methods that either require a geometric template of the object in a rest pose [27], or prior knowledge of the skeletal structure [23].”
Hand Gesture Recognition Based on Computer Vision: A Review of Techniques (2020, methodology): “[83] proposed a new system tracking human hand by combine surface model with new energy function which continuously optimized jointly over pose and correspondences, which can track the hand for several meter from the camera.”
InterHand2.6M: A Dataset and Baseline for 3D Interacting Hand Pose Estimation from a Single RGB Image (2020, unknown): “[35] proposed to perform joint optimization over both the hand model pose and the correspondences between observed data points and the hand model surface.”
Cross-Modal Deep Variational Hand Pose Estimation (2018, background): “The complex and dexterous articulation of the human hand is difﬁcult to model directly with geometric or physical constraints [18, 30, 33].”
Embodied hands (2017, methodology): “All above approaches model hands in isolation from the rest of the body, with the exception of [Melax et al. 2013; Taylor et al. 2016] that also model the forearm.”