Action Recognition

Human action recognition has become an important research area in the fields of computer vision, image processing, and human-machine or human-object interaction due to its large number of real time applications. Action recognition is the identification of different actions from video clips (an arrangement of 2D frames) where the action may be performed in the video. This is a general construction of image classification tasks to multiple frames and then collecting the predictions from each frame. Different approaches are proposed in literature to improve the accuracy in recognition. In this paper we proposed a deep learning based model for Recognition and the main focus is on the CNN model for image classification. The action videos are converted into frames and pre-processed before sending to our model for recognizing different actions accurately..

This technical report presents our approach “Knights” to solve the action recognition task on a small subset of Kinetics-400 i.e. Kinetics400ViPriors without using any extra-data. Our approach has 3 main components: state-of-the-art self-supervised pretraining, video transformer models, and optical flow modality. Along with the use of standard test-time augmentation, our proposed solution achieves 73% on Kinetics400ViPriors test set, which is the best among all of the other entries Visual Inductive Priors for Data-Efficient Computer Vision’s Action Recognition Challenge, ICCV 2021.

This paper summarizes the TinyAction challenge 1 which was organized in ActivityNet workshop at CVPR 2021. This challenge focuses on recognizing real-world low-resolution activities present in videos. Action recognition task is currently focused around classifying the actions from highquality videos where the actors and the action is clearly visible. While various approaches have been shown effective for recognition task in recent works, they often do not deal with videos of lower resolution where the action is happening in a tiny region. However, many real world security videos often have the actual action captured in a small resolution, making action recognition in a tiny region a challenging task. In this work, we propose a benchmark dataset, TinyVIRAT-v2 2, which is comprised of naturally occuring low-resolution actions. This is an extension of the TinyVIRAT dataset [7] and consists of actions with multiple labels. The videos are extracted from security videos which makes them r...

Human action recognition based on joints is a challenging task. The 3D positions of the tracked joints are very noisy if occlusions occur, which increases the intra-class variations in the actions. In this paper, we propose a novel approach to recognize human actions with weighted joint-based features. Previous work has focused on hand-tuned joint-based features, which are difficult and time-consuming to be extended to other modalities. In contrast, we compute the joint-based features using an unsupervised learning approach. To capture the intra-class variance, a multiple kernel learning approach is employed to learn the skeleton structure that combine these joints-base features. We test our algorithm on action application using Microsoft Research Action3D (MSRAction3D) dataset. Experimental evaluation shows that the proposed approach outperforms state-of-the-art action recognition algorithms on depth videos.

In this paper the use of sequential machines for recognizing actions taken by the objects detected by a general tracking algorithm is proposed. The system may deal with the uncertainty inherent in medium-level vision data. For this purpose, fuzzification of input data is performed. Besides, this transformation allows to manage data independently of the tracking application selected and enables adding characteristics of the analyzed scenario. The representation of actions by means of an automaton and the generation of the input symbols for finite automaton depending on the object and action compared are described. The output of the comparison process between an object and an action is a numerical value that represents the membership of the object to the action. This value is computed depending on how similar the object and the action are. The work concludes with the application of the proposed technique to identify the behavior of vehicles in road traffic scenes.

We present a Bag-of-Visual-and-Depth-Words (BoVDW) model for gesture recognition, an extension of the Bag-of-Visual-Words (BoVW) model, that benefits from the multimodal fusion of visual and depth features. State-of-the-art RGB and depth features, including a new proposed depth descriptor, are analysed and combined in a late fusion fashion. The method is integrated in a continuous gesture recognition pipeline, where Dynamic Time Warping (DTW) algorithm is used to perform prior segmentation of gestures ...

International Conference on Vision and Computational Intelligence (VCOI 2023) will
provide an excellent international forum for sharing knowledge and results in theory, methodology
and applications of Vision Computing and Computational Intelligence. The goal of this conference
is to bring together researchers and practitioners from academia and industry to focus on
understanding advances in vision computing & Computational Intelligence and establishing new
collaborations in these areas. Authors are solicited to contribute to the conference by submitting
articles that illustrate research results, projects, surveying works and industrial experiences that
describe significant advances in the areas of vision computing & Computational Intelligence.

International Conference on Vision and Computational Intelligence (VCOI 2023) will
provide an excellent international forum for sharing knowledge and results in theory, methodology
and applications of Vision Computing and Computational Intelligence. The goal of this conference
is to bring together researchers and practitioners from academia and industry to focus on
understanding advances in vision computing & Computational Intelligence and establishing new
collaborations in these areas. Authors are solicited to contribute to the conference by submitting
articles that illustrate research results, projects, surveying works and industrial experiences that
describe significant advances in the areas of vision computing & Computational Intelligence.

International Conference on Vision and Computational Intelligence (VCOI 2023) will provide an excellent international forum for sharing knowledge and results in theory, methodology and applications of Vision Computing and Computational Intelligence. The goal of this conference is to bring together researchers and practitioners from academia and industry to focus on understanding advances in vision computing & Computational Intelligence and establishing new collaborations in these areas. Authors are solicited to contribute to the conference by submitting articles that illustrate research results, projects, surveying works and industrial experiences that describe significant advances in the areas of vision computing & Computational Intelligence.

In the main text, both the proposed approaches have been described in detail. The methods have been extensively evaluated using three sets of multi-domain action datasets. In the main text, due to space constraint, only the results for UO (UCF50 and Olympic Sports) and KMS (KTH, MSR Action II and Sonycam) datasets were presented and discussed and the additional results have been included in the supplementary material. We start with the algorithmic details of Action Modeling on Latent Space (AMLS) approach. In the next section, we describe the Symmetrized KL Divergence measure and then discuss the KMS and UO datasets with few example images. In Section 5, we describe our third dataset collection (HU) comprising five common classes of HMDB51 and UCF50 and present its domain adaptation results. In the next section, we discuss the qualitative analysis of the results for the HU dataset and present some of the negative examples observed in our experiments. Finally, in Section 7, we discus...

Human action recognition has become an important research area in the fields of computer vision, image processing, and human-machine or human-object interaction due to its large number of real time applications. Action recognition is the identification of different actions from video clips (an arrangement of 2D frames) where the action may be performed in the video. This is a general construction of image classification tasks to multiple frames and then collecting the predictions from each frame. Different approaches are proposed in literature to improve the accuracy in recognition. In this paper we proposed a deep learning based model for Recognition and the main focus is on the CNN model for image classification. The action videos are converted into frames and pre-processed before sending to our model for recognizing different actions accurately..