2026-04-17T11:29:01Zhttps://recercat.cat/oai/request

oai:recercat.cat:2117/1175602025-07-22T17:18:16Zcom_2072_1033col_2072_452951

Online action detection Pan, Junting Giró Nieto, Xavier Chang, Shih-Fu Àrees temàtiques de la UPC::Enginyeria de la telecomunicació Machine learning Neural networks (Computer science) Video analysis Online action detection Convolutional neural networks Deep learning Generative adversarial networks Aprenentatge automàtic Xarxes neuronals (Informàtica) The details of the work will be defined when the student reaches his destination. In online detection, the objective is to detect the start of an action in a video stream as soon as it happens. It is an important yet challenging problem. In many realistic scenarios, we need to detect the action before the action is completed. For example, in the autonomous driving system, it is crucial to detect whether the pedestrian is crossing the street well in time in order to make a decision to stop or to reduce the velocity. Online action detection is a very challenging task in many aspects. It is very hard to predict the start of action for three reasons: First, the background is very diverse. Moreover, there is only a few action instance in a very long video. Last but not least, the model only observes part of the action to predict. To address those challenges, we propose a framework for online action detection and simulate experiments on a large-scale untrimmed video dataset. With the proposed method we have obtained very competitive performance. We also proposed a new evaluation metric for online detection models: Point mean Average Precision (Point mAP), a more appropriate metric than the existing evaluation metrics that have been designed for action detection in an offline setting. We have conducted experiments on THUMOS'14 dataset of video analysis where our proposed model achieved the state-of-the-art performance on the online action detection task. 2017-10 Master thesis https://hdl.handle.net/2117/117560 ETSETB-230.127224 eng S'autoritza la difusió de l'obra mitjançant la llicència Creative Commons o similar 'Reconeixement-NoComercial- SenseObraDerivada' http://creativecommons.org/licenses/by-nc-nd/3.0/es/ Open Access application/pdf Universitat Politècnica de Catalunya