Search

Human action recognition in video is one of the key problems in visual data interpretation. Despite intensive research, the recognition of actions with low inter-class variability remains a challenge. This paper presents a new Twin Spatio-Temporal Convolutional Neural Network (TSTCNN) for this purpose. When applied to table tennis, it is possible to detect and recognize 20 table tennis strokes. The model has been trained on a specific dataset, so called TTStroke-21, recorded in natural conditions at the Faculty of Sports of the University of Bordeaux. Our model takes as inputs an RGB image sequence and its computed Optical Flow. The proposed Twin architecture is a two stream network both comprising 3 spatio-temporal convolutional layers, followed by a fully connected layer where data are fused. Our method reaches an accuracy of 91.4% against 43.1% for our baseline, a Two-Stream Inflated 3D ConvNet (I3D).

This paper proposes a fusion method of modalities extracted from video through a three-stream network with spatio-temporal and temporal convolutions for fine-grained action classification in sport. It is applied to TTStroke-21 dataset which consists of untrimmed videos of table tennis games. The goal is to detect and classify table tennis strokes in the videos, the first step of a bigger scheme aiming at giving feedback to the players for improving their performance. The three modalities are raw RGB data, the computed optical flow and the estimated pose of the player. The network consists of three branches with attention blocks. Features are fused at the latest stage of the network using bilinear layers. Compared to previous approaches, the use of three modalities allows faster convergence and better performances on both tasks: classification of strokes with known temporal boundaries and joint segmentation and classification. The pose is also further investigated in order to offer richer feedback to the athletes., Comment: MMSports '21, October 20, 2021, Virtual Event,, Oct 2021, Chengdu, China

International audience; The paper addresses the problem of recognition of actions in video with low inter-class variability such as Table Tennis strokes. Two stream, "twin" convolutional neural networks are used with 3D convolutions both on RGB data and optical flow. Actions are recognized by classification of temporal windows. We introduce 3D attention modules and examine their impact on classification efficiency. In the context of the study of sportsmen performances, a corpus of the particular actions of table tennis strokes is considered. The use of attention blocks in the network speeds up the training step and improves the classification scores up to 5% with our twin model. We visualize the impact on the obtained features and notice correlation between attention and player movements and position. Score comparison of state-of-the-art action classification method and proposed approach with attentional blocks is performed on the corpus. Proposed model with attention blocks outperforms previous model without them and our baseline.

International audience; Detecting and classifying human actions in videos is oneof the current challenges in visual content analysis and mining. This paper presents a method for performing a fine-grained classification of sport actions using a Siamese Spatio-Temporal Convolutional Neural Network (SSTCNN) model.This model takes RGB images and Optical Flow field as in-put data. Our first contribution is the comparison of different Optical flow methods and a study of their influence on the classification score. We also present different normalization methods for the optical flow that drastically impact results,boosting performances from 44% to 74% of accuracy. Our second contribution is the detection and classification of actions in videos performed using a sliding temporal window.It leads to a satisfying score of 81.3% over the whole dataset TTStroke-21.

International audience; Human action recognition in videos is one of the key problems in visual data interpretation. Despite intensive research, the recognition of actions with low inter-class variability remains a challenge. To answer this problem, my thesis focus on fine-grained classification challenge using a Siamese Spatio-Temporal Convolutional Neural Network and apply it to a new dataset we have introduced TTStroke-21. Our model take as input data RGB images and Optical Flow and is able to reach an accuracy of 91.4% against 43.1% for our baseline on temporal segmented videos. Detection and classification in videos using a sliding temporal window leads to a score of 81.3% over the whole dataset.

International audience; Human action recognition in video is one of the key problems in visual data interpretation. Despite intensive research, the recognition of actions with low inter-class variability remains a challenge. This paper presents a new Siamese Spatio-Temporal Convolutional neural network (SSTC) for this purpose. When applied to table tennis, it is possible to detect and recognize 20 table tennis strokes. The model has been trained on a specific dataset, TTStroke-21, recorded in natural condition (markerless) at the Faculty of Sports of the University of Bordeaux. Our model takes as inputs a RGB image sequence and its computed Optical Flow. After 3 spatio-temporal convolutions, data are fused in a fully connected layer of a proposed siamese network architecture. Our method reaches an accuracy of 91.4% against 43.1% for our baseline.

The research context of this article is the recognition and description of dynamic textures. In image processing, the wavelet transform has been successfully used for characterizing static textures. To our best knowledge, only two works are using spatio-temporal multiscale decomposition based on the tensor product for dynamic texture recognition. One contribution of this article is to analyze and compare the ability of the 2D+T curvelet transform, a geometric multiscale decomposition, for characterizing dynamic textures in image sequences. Two approaches using the 2D+T curvelet transform are presented and compared using three new large databases. A second contribution is the construction of these three publicly available benchmarks of increasing complexity. Existing benchmarks are either too small not available or not always constructed using a reference database. Feature vectors used for recognition are described as well as their relevance, and performances of the different methods are discussed. Finally, future prospects are exposed.

International audience; The research context of this work is dynamic texture analysis and characterization. Many dynamic textures can be modeled as large scale propagating wavefronts and local oscillating phenomena. After introducing a formal model for dynamic textures, the Morphological Component Analysis (MCA) approach with a well chosen dictionary is used to retrieve the components of dynamic textures. We define two new strategies for adaptive thresholding in the MCA framework, which greatly reduce the computation time when applied on videos. Tests on real image sequences illustrate the efficiency of the proposed method. An application to global motion estimation is proposed and future prospects are finally exposed.

This paper presents an original approach for recognizing human activities in video sequences. A human activity is seen as a temporal sequence of elementary action probabilities. Actions are first generically learned using a robust action recognition method based on optical flow estimation and a cross-dataset training process. Activities are then projected as trajectories on the semantic simplex in order to be characterized and discriminated. A new trajectory attribute based on the total curvature Fourier descriptor is introduced. This attribute takes into account the induced geometry of the simplex manifold. Experiments on labelled datasets of human activities prove the efficiency of the proposed method for discriminating complex actions.

International audience; This paper focuses on human action recognition in video sequences. A method based on the optical flow estimation is presented, where critical points of the flow field are extracted. Multi-scale trajectories are generated from those points and are characterized in the frequency domain. Finally, a sequence is described by fusing this frequency information with motion orientation and shape information. Experiments show that this method has recognition rates among the highest in the state of the art on the KTH dataset. Contrary to recent dense sampling strategies, the proposed method only requires critical points of motion flow field, thus permitting a lower computation time and a better sequence description. Results and perspectives are then discussed.

8 pages; International audience; The aim of this paper is to construct a model which decomposes a 3D image into two components: the fi rst one containing the geometrical structure of the image, the second one containing the noise. The proposed method is based on a second order variational model and an undecimated wavelet thresholding operator. The numerical implementation is described, and some experiments for denoising a 3D MRI imageare successfully performed. Future prospects are finally exposed.

The research context of this work is dynamic texture analysis and characterization. A dynamic texture can be described as a time-varying phenomenon with a certain repetitiveness in both space and time.

8 pages; International audience; This paper presents four spatio-temporal wavelet decompositions for characterizing dynamic textures. The main goal of this work is to compare the influence of spatial and temporal variables in the wavelet decomposition scheme. Its novelty is to establish a comparison between the only existing method and three other spatio-temporal decompositions. The four decomposition schemes are presented and successfully applied on a large dynamic texture database. Construction of feature descriptors are tackled as well their relevance, and performances of the methods are discussed. Finally, future prospects are exposed.