Searching for variable-speed motions in long sequences of motion capture data

Sedmidubský, Jan; Eliáš, Petr; Zezula, Pavel

doi:10.1016/j.is.2018.04.002

Cited by 20 publications

(21 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By comparing the recognition rates of high‐level descriptors and PMI high‐level descriptors under the same dictionary size, the conclusion is drawn that our proposed algorithm is superior to the semantic similarity calculation method based on, and the recognition rate is increased by 0.67% on average. In addition, by comparing the recognition rates of high‐level descriptors and PMI high‐level descriptors [19] under the same dictionary size, the same conclusion is drawn that the near‐semantic visual dictionary generation algorithm proposed in this chapter is superior to the PPM‐based high‐level descriptors in recognition rate, which increases the average recognition rate by 2.09%.…”

Section: Simulation and Results Analysismentioning

confidence: 64%

Multi‐dimensional data modelling of video image action recognition and motion capture in deep learning framework

Gao

Zhao

Chen

2020

IET Image Processing

View full text Add to dashboard Cite

Section: Simulation and Results Analysismentioning

confidence: 64%

Multi‐dimensional data modelling of video image action recognition and motion capture in deep learning framework

Gao

Zhao

Chen

2020

IET Image Processing

View full text Add to dashboard Cite

“…Most existing works that process continuous 3D skeleton sequences in an unsupervised way focus on subsequence search [18], unsupervised segmentation [8], or anticipating future actions based on the past-to-current data [4]. In [18], the continuous sequences are synthetically partitioned into a lot of overlapping and variable-size segments that are represented by 4, 096D deep features. However, indexing a large number of such very high-dimensional features is costly.…”

Section: Related Work and Our Contributionsmentioning

confidence: 99%

Motion Words: A Text-Like Representation of 3D Skeleton Sequences

Sedmidubský

Budikova

Dohnal

et al. 2020

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

There is a growing amount of human motion data captured as a continuous 3D skeleton sequence without any information about its semantic partitioning. To make such unsegmented and unlabeled data efficiently accessible, we propose to transform them into a text-like representation and employ well-known text retrieval models. Specifically, we partition each motion synthetically into a sequence of short segments and quantize the segments into motion words, i.e. compact features with similar characteristics as words in text documents. We introduce several quantization techniques for building motion-word vocabularies and propose application-independent criteria for assessing the vocabulary quality. We verify these criteria on two real-life application scenarios.

show abstract

“…Just as content-based image retrieval (CBIR) [ 17 ], feature extraction and similarity measure are two most important parts in CBMR. Although the query example of CBIR is commonly accepted as a single image, query example in the QBE systems of CBMR can be found in various forms, such as a single 3D pose [ 3 ], strictly aligned motion clip [ 6 , 7 ], slightly misaligned motion clip [ 18 ], depending on their different assumptions on the boundaries of motion clip. It can be seen that most of the current CBMR methods impose additional constraints on query examples, regardless the fact that obtaining 3D query examples of motion is hard for users.…”

Section: Related Workmentioning

confidence: 99%

Visual Browse and Exploration in Motion Capture Data with Phylogenetic Tree of Context-Aware Poses

Chen

Zhao

Luo

2020

Sensors

View full text Add to dashboard Cite

Visual browse and exploration in motion capture data take resource acquisition as a human–computer interaction problem, and it is an essential approach for target motion search. This paper presents a progressive schema which starts from pose browse, then locates the interesting region and then switches to online relevant motion exploration. It mainly addresses three core issues. First, to alleviate the contradiction between the limited visual space and ever-increasing size of real-world database, it applies affinity propagation to numerical similarity measure of pose to perform data abstraction and obtains representative poses of clusters. Second, to construct a meaningful neighborhood for user browsing, it further merges logical similarity measures of pose with the weight quartets and casts the isolated representative poses into a structure of phylogenetic tree. Third, to support online motion exploration including motion ranking and clustering, a biLSTM-based auto-encoder is proposed to encode the high-dimensional pose context into compact latent space. Experimental results on CMU’s motion capture data verify the effectiveness of the proposed method.

show abstract

Searching for variable-speed motions in long sequences of motion capture data

Cited by 20 publications

References 21 publications

Multi‐dimensional data modelling of video image action recognition and motion capture in deep learning framework

Multi‐dimensional data modelling of video image action recognition and motion capture in deep learning framework

Motion Words: A Text-Like Representation of 3D Skeleton Sequences

Visual Browse and Exploration in Motion Capture Data with Phylogenetic Tree of Context-Aware Poses

Contact Info

Product

Resources

About