Recognizing Actions from Depth Cameras as Weakly Aligned Multi-part Bag-of-Poses

By Lorenzo Seidenari, Vincenzo Varano, Stefano Berretti, A. del Bimbo, Pietro Pala. 2013

252 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Control and Systems Engineering

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

Citations in the corpus

Space-Time Representation of People Based on 3D Skeletal Data: A Review (2016, methodology): “Many other datasets captured using a Kinect v1 camera were also released to the public, including the MSR Daily Activity 3D [121], MSR Action Pairs [113], Online RGBD Action (ORGBD) [107], UTKinect-Action [124], Florence 3D-Action [118], CMU-MAD [104], UTD-MHAD [103], G3D/G3Di [105], [119], SPHERE [”
When Kernel Methods Meet Feature Learning: Log-Covariance Network for Action Recognition From Skeletal Data (2017, positive): “Namely, we evaluated on MSR-Action3D [20], MSR-Action-Pairs (MSR-pairs) [27], Gaming-3D (G3D) [2], Florence3D [31], UTKinect [40], MSRC-Kinect12 [8], HDM-05 [25] and NTU RGB+D [32].”
Tensor Representations via Kernel Linearization for Action Recognition from 3D Skeletons (2016, methodology): “Florence3D-Action [10] dataset consists of 9 actions performed two to three times by 10 different subjects.”
Human Action Recognition by Shape Analysis of Motion Trajectories on Riemannian Manifold (2014, methodology): “We compare our method with the one presented in [39].”
WALKING WALKing walking: Action Recognition from Action Echoes (2017, methodology): “[Seidenari et al. , 2013] Captured by using Kinect sensor, it contains 3D locations of 15 joints for 9 actions performed 2-3 times by 10 different subjects, giving a total of 215 sequences.”
Mining 3D Key-Pose-Motifs for Action Recognition (2016, methodology): “the MSR-Action3D dataset [11], the UTKinect dataset [30], the MSR Daily Activity3D dataset [27] and the Florence dataset [17].”
Evaluating spatiotemporal interest point features for depth-based action recognition (2014, unknown): “NBNN + parts + time [37] 70.”
Super Normal Vector for Activity Recognition Using Depth Sequences (2014, methodology): “The methods [10, 14, 21, 24, 26] based on either motion or shape information alone are significantly inferior to our method and the ones [12, 21] that jointly model the two cues.”
RGBD Datasets: Past, Present and Future (2016, background): “Florence 3D actions dataset [88] 10 9 215 X e.”
Evaluation of video activity localizations integrating quality and quantity measurements (2014, methodology): “Basic actions such as jumping, hand clapping, and stand up, are handled in the Berkeley Multimodal Human Action Database (MHAD) [45] as well as in the Florence3D Dataset [46].”
Vision-based human activity recognition: a survey (2020, methodology): “The authors of [198] use three datasets to test the performance of their approach; 3D MSR Action [90], UT Kinect [220] and 3D Florence Actions [169].”
A discussion on the validation tests employed to compare human action recognition methods using the MSR Action3D dataset (2014, unknown): “Florence 3D Action [17] 9 10 2 or 3 215 2 2012”
Learning features combination for human action recognition from skeleton sequences (2017, methodology): “Similarly to the previews experiment, we evaluate the performance of our method on the Florence 3D Actins dataset using the LOOCV approach, as suggested by the authors [18].”
Spatiotemporal representation of 3D skeleton joints-based action recognition using modified spherical harmonics (2016, methodology): “More-over, in our method, we perform the evaluation on a variety of datasets compared to many approaches ((Xia et al., 2012), (Yang and Tian, 2012), and (Seidenari et al., 2013)).”
Elastic functional coding of human actions: From vector-fields to latent variables (2015, background): “[31], and Florence3D [18] datasets obtained with mfPCA.”
Spatio-Temporal Graph Convolution for Skeleton Based Action Recognition (2018, methodology): “We conduct experiments on four public skeleton based action datasets: Florence 3D (Seidenari et al. 2013), HDM05 (M¨uller et al. 2007), Large Scale Combined dataset (Zhang et al. 2016)”