An Efficient Dense and Scale-Invariant Spatio-Temporal Interest Point Detector

By Geert Willems, Tinne Tuytelaars, Luc Van Gool. 2008

1,037 citations recorded by the source corpus.

Subfields: 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 Computer Vision and Pattern Recognition

Citations in the corpus

Histogram of Oriented Principal Components for Cross-View Action Recognition (2014, methodology): “Experimental results show that our method is able to achieve significantly better accuracy than current state-of-the-art methods for cross-view action recognition and is robust to action speed variations.”
Action recognition with trajectory-pooled deep-convolutional descriptors (2015, methodology): “Meanwhile several local descriptors have been proposed to represent the 3D volumes extracted around these interest points, such as Histogram of Gradient (HOG), Histogram of Optical Flow (HOF) [17], 3D Histogram of Gradient (HOG3D) [13], and Extended SURF (ESURF) [39].”
Bag of visual words and fusion methods for action recognition: Comprehensive study and good practice (2014, neutral): “The main contents of LTC are manifold approximation and intrinsic dimensionality estimation.”
Temporal Segment Networks for Action Recognition in Videos (2017, methodology): “…points or trajectories, and a histogram descriptor is computed to capture the appearance and motion information, such as Histogram of Gradient and Histogram of Flow (HOG/HOF) [39], Histogram of Motion Boundary (MBH) [38], 3D Histogram of Gradient (HOG3D) [40], Extended SURF (ESURF) [36], and so on.”
Histograms of oriented optical flow and Binet-Cauchy kernels on nonlinear dynamical systems for the recognition of human actions (2009, background): “Local approaches use local spatiotemporal features to represent human activity in a video, e.g. [17] [10] [30].”
Exemplar-based Action Recognition in Video (2009, methodology): “Roughly speaking, methods for object detection (localization) can be divided into sliding window-based approaches and constellation-based approaches, where the former are clearly predominant (see e.g. recent editions of the Pascal VOC challenge [6]).”
Spatial-Aware Object Embeddings for Zero-Shot Localization and Classification of Actions (2017, neutral): “Such representations include local spatio-temporal interest points and features [26, 31, 52] and local trajectories [1, 50], typically aggregated into VLAD or Fisher vector representations [38, 39].”
Person Re-identification by Video Ranking (2014, background): “One common representation is constructed based on space-time interest points [26,10,46,5].”
MARS: A Video Benchmark for Large-Scale Person Re-Identification (2016, background): “Space-time features can be extracted based on the space-time interest points [9, 21, 37].”
A temporal Bayesian model for classifying, detecting and localizing activities in video sequences (2012, background): “Spatiotemporal interest points were earlier introduced by Laptev and Linde [14] and since then other interest-point-based detectors such as those based on spatio-temporal Hessian matrix [26] and Gabor filters [3, 6] have been proposed.”
A Review on Human Activity Recognition Using Vision-Based Method (2017, methodology): “Therefore, these methods are restricted to the detection of a small number of points, or limited to low-resolution videos [39].”
Harnessing Lab Knowledge for Real-World Action Recognition (2014, methodology): “There are some other feature detectors such as Hessian (Willems et al. 2008), dense sampling and feature descriptors such as HOG3D (Kläser et al. 2008) and the extended SURF (ESURF) (Willems et al. 2008) etc. which are widely used for video representation.”
Action Recognition by Learning Deep Multi-Granular Spatio-Temporal Video Representation (2016, background): “The global color moment feature [34], Histogram of Gradient (HOG) and Histogram of Optical Flow (HOF) [15], 3D Histogram of Gradient (HOG3D) [10], SIFT-3D [21], Extended SURF [31], and Cuboids [2] are good descriptors as the local spatio-temporal features.”
Exploiting Image-trained CNN Architectures for Unconstrained Video Classification (2015, background): “, [6, 17, 18, 35, 38]) or along estimated motion trajectories (e.”