Predicting missing markers in human motion capture using l1‐sparse representation

Abstract

Missing marker problem is very common in human motion capture. In contrast to most current methods which handle this problem based on trying to learn a reliable predictor from the observations, we consider it from the perspective of sparse representation and propose a novel method which is named l1-sparse representation of missing markers prediction (L1-SRMMP). We assume that the incomplete pose can be represented by a linear combination of a few poses from the training set and the representation is sparse. Therefore, we cast the predicting missing markers as finding a sparse representation of the observable data of the incomplete pose, and then we use it to predict the missing data. In order to get a sparse representation, we employ l1-norm in our objective function. Moreover, we propose presentation coefficient weighted update (PCWU) algorithm to mitigate the limited capacity problem of the training set. Experimental results demonstrate the effectiveness and efficiency of our method to predict the missing markers in human motion capture.

Human motion data refinement unitizing structural sparsity and spatial-temporal information

“…Two motion sequences from each category are randomly selected as testing set while others are used for training. Most [24], [32], a linear dynamic system(LDS) based method; SRMMP [9], a sparse coding based method for predicting missing markers; SVT [21], a matrix completion based method.…”

“…[4]- [6], [8], [9], [20]. For example, Lou and Chai [4] have proposed an example based approach to learn a series of spatial-temporal filter bases from pre-captured motion data and use them along with robust statistics techniques to fill in the missing values of motion capture data.…”

“…For example, Lou and Chai [4] have proposed an example based approach to learn a series of spatial-temporal filter bases from pre-captured motion data and use them along with robust statistics techniques to fill in the missing values of motion capture data. In Xiao et al's work [6], [9], they have formulated the predicting missing marker problem as finding spare representation of imperfect pose. They succussed in introducing 1 sparse representation to solve predicting missing marker of motion data.…”

“…In order to better exploit the spatial-temporal relationship, many researches have applied the partial human model while processing motion data [5], [6], [25], [27], [28] and achieved expressive performance. Instead of using whole body model [8], [9], We choose partial human model in this work and divide the whole body into 5 parts [5], [6], which is…”

“…[4] and Xiao et.al. [9], the spatial temporal and kinematic information of human motion haven't been well exploited while training the motion dictionaries. The artifact e.g.…”