Human Activity Detection from RGBD Images

By Jaeyong Sung, Colin Ponce, Bart Selman, Ashutosh Saxena. 2011

355 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, 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 Computer Vision and Pattern Recognition

In 3D Vision & Pose

Citations in the corpus

NTU RGB+D: A Large Scale Dataset for 3D Human Activity Analysis (2016, background): “CAD-60 [34] and CAD-120 [18] contain RGB, depth, and skeleton data of human actions.”
NTU RGB+D 120: A Large-Scale Benchmark for 3D Human Activity Understanding (2019, background): “The CAD-60 [12] and CAD-120 [17] datasets contain RGB, depth, and skeleton data of human actions.”
Co-Occurrence Feature Learning for Skeleton Based Action Recognition Using Regularized Deep LSTM Networks (2016, methodology): “The other is to explore temporal dependencies of the inter-frame content for action dynamics modeling, using hierarchical maximum entropy Markov model (Sung et al. 2011), hidden Markov model (Xia, Chen, and Aggarwal 2012) or Conditional Random Fields (Sminchisescu et al. 2005).”
RGB-D-based action recognition datasets: A survey (2016, background): “CAD-60 dataset [12]( http://pr.cs.cornell.edu/humanactivities/data.php ) was captured by Cornell University in 2011, motivated by the fact that true daily activities rarely occur in structured environments.”
Space-Time Representation of People Based on 3D Skeletal Data: A Review (2016, methodology): “A similar approach was also introduced in [148] based on the orientation matrix.”
Learning human activities and object affordances from RGB-D videos (2012, background): “Recently, with the availability of inexpensive RGB-D sensors, some works (Li et al., 2010; Ni et al., 2011; Sung et al., 2011; Zhang and Parker, 2011; Sung et al., 2012) consider detecting human activities from RGB-D videos.”
Modeling Temporal Dynamics and Spatial Configurations of Actions Using Two-Stream Recurrent Neural Networks (2017, background): “Sung et al. [41] compute the joint’s rotation matrix w.r.t. the person’s torso, hand position w.r.t. the torso and joint rotation motion as features.”
PKU-MMD: A Large Scale Benchmark for Continuous Multi-Modal Human Action Understanding (2017, background): “Compared to CAD-60, CAD-120 provides extra labels of temporal locations.”
Unsupervised Learning of Long-Term Motion Dynamics for Videos (2017, background): “However, these datasets were often constructed under various constraints like single-view, single background, or with very few subjects.”
ScanNet: Richly-Annotated 3D Reconstructions of Indoor Scenes (2017, background): “A large number of RGB-D datasets have been captured and made publicly available for training and benchmarking [53, 32, 48, 62, 76, 80, 71, 4, 55, 78, 15, 52, 1, 65, 29, 49, 21, 46, 41, 89, 77, 58, 69, 90, 34, 16, 33, 54, 38, 28, 67, 50, 43, 92, 72, 9, 31, 82, 68, 30, 3, 10, 75, 2].1 These datasets…”
Anticipating Human Activities Using Object Affordances for Reactive Robotic Response (2013, background): “Recent availability of inexpensive RGB-D sensors has enabled significant improvement in scene modeling [30], [31] and estimation of human poses [32], [33].”
Human activities recognition using depth images (2013, neutral): “Copyright 2013 ACM 978-1-4503-2404-5/13/10 ...$15.00.”
Human gesture recognition through a Kinect sensor (2012, background): “Chinese Academy of Sciences, Shenzhen, Guangdong 518055, China. ys.ou@siat.ac.cn”
Video-Based Human Behavior Understanding: A Survey (2013, methodology): “The CAD-60 and CAD-120 datasets consider most environments of a regular household (office, kitchen, bedroom, bathroom), with three to four common activities identified in each location with about 45 s of data for each activity and each person.”
Human Daily Action Analysis with Multi-view and Color-Depth Data (2012, background): “[9] construct a RGBD dataset for human activity detection and recognition in the living environment, and the RGBD-HuDaAct [10] is aiming to encourage application of assisted living in health-care.”
Jointly Learning Heterogeneous Features for RGB-D Activity Recognition (2015, methodology): “This public dataset consists of 68 video clips captured by Microsoft Kinect device [33].”
Learning Object Arrangements in 3D Scenes using Human Context (2012, methodology): “We extract human skeletons from the Kinect RGB-D dataset (Sung et al., 2012), which contains activities performed by four subjects.”
Graph Distillation for Action Detection with Privileged Modalities (2017, background): “The success in RGB videos has given rise to a series of studies on action recognition in multimodal videos [10,20,22,25,51,55].”