Action Recognition with Stacked Fisher Vectors

By Xiaojiang Peng, Changqing Zou, Yu Qiao, Qiang Peng. 2014

388 citations recorded by the source corpus.

Subfields: Artificial Intelligence, 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 Artificial Intelligence

In Biomedical Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Scene Flow to Action Map: A New Representation for RGB-D Based Action Recognition with Convolutional Neural Networks (2017, background): “Two kinds of motion representations can be identified: Lagrangian motion [48, 18, 31, 50, 30, 56] and Eulerian motion [2, 25, 60, 51, 52, 1].”
Beyond Gaussian Pyramid: Multi-skip Feature Stacking for action recognition (2014, methodology): “[24, 25] further improved the performance of Dense Trajectory by increasing the codebook sizes, fusing multiple coding methods and adding a stacked Fisher Vector.”
Actions ~ Transformations (2015, background): “8% [32]).”
Action Recognition using Visual Attention (2015, neutral): “0 Traditional+Stacked Fisher Vectors (Peng et al., 2014) 66.”
Long-Term Temporal Convolutions for Action Recognition (2016, neutral): “F101 and RGB scores are obtained using C3D feature extractor. Long-term Temporal Convolutions 13 Method UCF101 HMDB51 [5] IDT+FV 85.9 57.2 Hand- [29] IDT+HSV 87.9 61.1 crafted [30] IDT+MIFS 89.1 65.1 [31] IDT+SFV - 66.8 [12] Slow fusion (from scratch) 41.3 - [13] C3D (from scratch) 441 - [12] Slow f”
Fisher Vectors Derived from Hybrid Gaussian-Laplacian Mixture Models for Image Annotation (2014, neutral): “However, the leading pooling technique in recent years is the Fisher Vector [32], which provided state-of-the-art results on many different applications [38, 31, 2, 33].”
VideoLSTM convolves, attends and flows for action recognition (2016, neutral): “We reckon and value the impact of traditional video representations, e.g., [3,4,5,19,20,21,22], and recent mixtures of shallow and deep encodings e.g., [23,24].”
When Video Classification Meets Incremental Classes (2021, background): “With the rise of deep learning, CNNs dominated the video understanding literature recently [3, 21, 27, 30, 36, 44, 54, 56], instead of hand-crafted features [1, 26, 35, 46, 47].”
Asynchronous Temporal Fields for Action Recognition (2016, background): “Understanding activities and actions has an extensive history [32, 59, 22, 17, 23, 2, 26, 56, 29, 21].”
Learnable pooling with Context Gating for video classification (2017, background): “However, the codebook can also be learned in a discriminative manner [34], [37], [38] or the entire encoding module can be included within the convolutional neural network architecture and trained in the end-to-end manner [23].”
A Comprehensive Survey of Vision-Based Human Action Recognition Methods (2019, methodology): “[64] used stacked Fisher vectors to represent human actions based on iDT features, resulting in one of the more successful methods for improving iDT.”
Hierarchical Attention Network for Action Recognition in Videos (2016, methodology): “These results further support that (1) the hierarchical structure in HAN is important because it can model the frame transitions as well [19] 85.9% 57.2% iDT + Stack Fish Vector [13] N/A 66.8% spatial-temporal CNN [10] 65.4% N/A two stream CNN [16] 88.0% 59.4% two stream CNN+LSTM [22] 88.6% N/A two ”
Human Action Recognition in Unconstrained Videos by Explicit Motion Modeling (2015, methodology): “[67] used a new Fisher vector encoding method and achieved very strong results.”
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].”