Skeleton-aware networks for deep motion retargeting

Abstract

We introduce a novel deep learning framework for data-driven motion retargeting between skeletons, which may have different structure, yet corresponding to homeomorphic graphs. Importantly, our approach learns how to retarget without requiring any explicit pairing between the motions in the training set. We leverage the fact that different homeomorphic skeletons may be reduced to a common primal skeleton by a sequence of edge merging operations, which we refer to as skeletal pooling. Thus, our main technical contribution is the introduction of novel differentiable convolution, pooling, and unpooling operators. These operators are skeleton-aware , meaning that they explicitly account for the skeleton's hierarchical structure and joint adjacency, and together they serve to transform the original motion into a collection of deep temporal features associated with the joints of the primal skeleton. In other words, our operators form the building blocks of a new deep motion processing framework that embeds the motion into a common latent space, shared by a collection of homeomorphic skeletons. Thus, retargeting can be achieved simply by encoding to, and decoding from this latent space. Our experiments show the effectiveness of our framework for motion retargeting, as well as motion processing in general, compared to existing approaches. Our approach is also quantitatively evaluated on a synthetic dataset that contains pairs of motions applied to different skeletons. To the best of our knowledge, our method is the first to perform retargeting between skeletons with differently sampled kinematic chains, without any paired examples.

Smoothing Skeleton Avatar Visualizations Using Signal Processing Technology

“…Finally, we acknowledge the work on estimating volume data-as opposed to key-point-based stick figures-of human poses and movements from images and videos, such as suggested by Kocabas et al [20]. We also acknowledge the work on creating free animations-as opposed to scientific visualizations-let it be volume animations [43] or stick figure animations [1]. All these somewhat related approaches are outside the scope of this paper.…”

“…The positions of key-points are then estimated using statistics and artificial intelligence (AI)-based software. Commodity 3D SAT systems including Microsoft' Kinect, 1 Orbbec's Astra Mini, 2 and Intel's Real Sense 3 use a video camera and a infrared depth sensor. Commodity 2D SAT systems only use a regular video camera.…”

Interactive multi‐character motion retargeting

“…Villegas et al 19 proposed an unsupervised motion retargeting system established by neural kinematic networks. Aberman et al 20 introduced convolutional neural networks-based motion retargeting system with skeleton convolution and pooling operation. These approaches produce a natural retargeted character animation.…”

OUTPUT: Choreographed and Reconfigured Human and Industrial Robot Bodies Across Artistic Modalities

“…Choreographers, researchers, and engineers alike have employed notation (such as Labanotation, Eshkol-Wachmann Notation, and Action Stroke Notation ( Eshkol et al, 1970 ; Hutchinson et al, 1977 ; Badler and Smoliar, 1979 ; Cooper, 1997 ) and abstraction (such as stick figures or animations ( Marr and Nishihara, 1978 ; Badler and Smoliar, 1979 ) to capture and demonstrate motion sequences. Human movement has been utilized as source material for humanoid robots with differing kinematic structures via mappings ( Do et al, 2008 ) and deep learning techniques ( Aberman et al, 2020 ). Industrial robots have appeared in live performances and installations.…”