STOP: Space-Time Occupancy Patterns for 3D Action Recognition from Depth Map Sequences

By Antônio Wilson Vieira, Erickson R. Nascimento, Gabriel L. Oliveira, Zicheng Liu, Mario Fernando Montenegro Campos. 2012

293 citations recorded by the source corpus.

Subfields: Biomedical Engineering, Computer Vision and Pattern Recognition, Endocrinology, Diabetes and Metabolism, 3D Vision & Pose

How researchers see this paper

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

In Biomedical Engineering

In Computer Vision and Pattern Recognition

In Endocrinology, Diabetes and Metabolism

In 3D Vision & Pose

Citations in the corpus

Space-Time Representation of People Based on 3D Skeletal Data: A Review (2016, background): “We also provide a brief overview of skeleton acquisition devices and construction methods, enlist a number of benchmark datasets with skeleton data, and discuss potential future research directions.”
Histogram of Oriented Principal Components for Cross-View Action Recognition (2014, methodology): “Moreover, we introduce a new UWA3D Multiview Activity II dataset in addition to [37] which contains 30 actions performed by 10 subjects from four different views.”
HOPC: Histogram of Oriented Principal Components of 3D Pointclouds for Action Recognition (2014, methodology): “Action recognition methods based on depth maps can be divided into holistic [19,21,13,38,29] and local approaches [32,35,11,31].”
Learning and Refining of Privileged Information-Based RNNs for Action Recognition from Depth Sequences (2017, background): “Action recognition from depth sequences [57, 34, 29, 44, 49] has attracted significant interest recently due to the emergence of low-cost depth sensors.”
Action Classification with Locality-Constrained Linear Coding (2014, methodology): “[13] computed silhouettes in 3D by using the space-time occupancy patterns.”
Evaluating spatiotemporal interest point features for depth-based action recognition (2014, unknown): “[15] proposed the Space–Time Occupancy Patterns (STOP) to represent sequences of depth maps.”
Human Action Recognition by Shape Analysis of Motion Trajectories on Riemannian Manifold (2014, background): “For example, in [19] a Local Occupancy Pattern around each 3D joint is proposed.”
HON4D: Histogram of Oriented 4D Normals for Activity Recognition from Depth Sequences (2013, methodology): “The HON4D descriptor discussed earlier is a holistic feature, similar in that to [26] and [21].”
Real-Time Gesture Recognition from Depth Data through Key Poses Learning and Decision Forests (2012, methodology): “The same dataset and subsets are used to compare with [10], [15] and our method, although we extract skeletons from the original depth maps.”
3D skeleton-based human action classification: A survey (2016, background): “In [133], space and time axes are divided in cells and space – time occupancy patterns are computed to represent depth sequences.”
Spatio-temporal Depth Cuboid Similarity Feature for Activity Recognition Using Depth Camera (2013, methodology): “Our algorithm outperforms the algorithms based on 3D silhouette features [13], skeletal joint features [30, 27] and local occupancy patterns[25, 26].”
A Survey on Human Motion Analysis from Depth Data (2013, methodology): “The authors compared their results against those obtained from [8] and [13] using the MSR Action3D Dataset [8].”
Human Action Recognition From Various Data Modalities: A Review (2020, methodology): “Similar to the STOP descriptors [255], the descriptors designed in [291] captured a point cloud structure through a modified 3D regular grid with corresponding cell space occupancy information, which were then used for HAR.”
Accurate 3D action recognition using learning on the Grassmann manifold (2015, methodology): “In the work proposed by Vieira et al. [31], each depth map sequence is represented as a 4D grid by dividing the space and time axes into multiple segments in order to extract SpatioTempo-182 ral Occupancy Pattern features (STOP).”
Range-Sample Depth Feature for Action Recognition (2014, background): “Occupancy descriptor [17, 16] divides a spatio-temporal region into a 3D grid where the number of occupied pixels is counted in each cell.”