ViVoLab and CVLab - MediaEval 2014: Violent Scenes Detection Affect Task
نویسندگان
چکیده
This paper describes the ViVoLab/CVLab system to provide segments of violent scenes from Hollywood movies and “wild-user” videos from the internet. We propose a system based on a fusion of acoustic features, audio concepts and video features. Our joint audio-visual approach achieves MAP2014 values of 17.81% and 43.03%, for the main task and the generalization task, respectively.
منابع مشابه
The MediaEval 2014 Affect Task: Violent Scenes Detection
This paper provides a description of the MediaEval 2014 Affect Task: Violent Scenes Detection, which is running for the fourth year. The task originates from a use case at Technicolorthat aims to help users find suitable contents from a movie database. We provide insights on the use case, task challenges, data set and ground truth, required and optional participant runs and evaluation metrics.
متن کاملThe MediaEval 2011 Affect Task: Violent Scenes Detection in Hollywood movies
This paper provides a description of the MediaEval 2011 Affect Task: Violent Scenes Detection. This task derives directly from a Technicolor use case which aims at easing a user’s selection process from a movie database. This task will therefore apply to movie content. We provide some insight into the Technicolor use case, before giving details on the task itself. Dataset, annotations, and eval...
متن کاملThe MediaEval 2012 Affect Task: Violent Scenes Detection
This paper provides a description of the MediaEval 2012 Affect Task: Violent Scenes Detection. This task derives directly from a Technicolor use case which aims at easing a user’s selection process from a movie database. This task will therefore apply to movie content. We provide some insight into the Technicolor use case, before giving details on the task itself. Dataset, annotations, and eval...
متن کاملFAR at MediaEval 2014 Violent Scenes Detection: A Concept-based Fusion Approach
The MediaEval 2014 Violent Scenes Detection task challenged participants to automatically find violent scenes in a set of videos. We propose to first predict a set of midlevel concepts from low-level visual and auditory features, then fuse the concept predictions and features to detect violent content. With the objective of obtaining a higly generic approach, we deliberately restrict ourselves ...
متن کاملNII-UIT at MediaEval 2013 Violent Scenes Detection Affect Task
We present a comprehensive evaluation of shot-based visual and audio features for MediaEval 2013 Violent Scenes Detection Affect Task. To obtain visual features, we use global features, local SIFT features and motion features. For audio features, the popular MFCC is employed. Besides that, we also evaluate the performance of mid-level features which is constructed using visual concepts. We comb...
متن کامل