Hollywood 3D: Recognizing Actions in 3D Natural Scenes

By Simon Hadfield, Richard Bowden. 2013

106 citations recorded by the source corpus.

Subfields: Biomedical Engineering, Computer Vision and Pattern Recognition

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

Citations in the corpus

Online RGB-D gesture recognition with extreme learning machines (2013, methodology): “In [8], actions from the database with 14 classes are classified by using both the RGB and depth data.”
Ieee Transactions on Pattern Analysis and Machine Inteligence 1 Super Normal Vector for Human Activity Recognition with Depth Cameras (unknown): “Hadfield et al. [10] extended the detection algorithms of Harris corners, Hessian points, and separable filters to the 3.5D and 4D for depth videos.”
Super Normal Vector for Activity Recognition Using Depth Sequences (2014, methodology): “[4] extended the detection algorithms of Harris corners, Hessian points, and separable filters to the 3.”
Bilinear heterogeneous information machine for RGB-D action recognition (2015, methodology): “This hinders the application of feature extraction methods such as surface normal [25, 13] and spatiotemporal interest points [23, 5] in these regions.”
Unsupervised learning of depth and motion (2013, methodology): “We evaluate and compare several variations of this approach on the Hollywood3D activity recognition dataset [6], and we demonstrate that it improves significantly upon the state-of-the-art, using a minimum of hand engineering.”
3D trajectories for action recognition (2014, background): “Recently many dedicated depth map descriptors have been proposed [3], [4], [5], but they are still limited in terms of high order search space and sensitivity to depth map noise or missing measurements.”
3D skeleton-based human action classification: A survey (2016, background): “Therefore, the table does not include datasets such as Hollywood 3D [146], which provides only depth data, or Human3.”