“…Face recognition could be used for video content description, indexing, and retrieval [25,96]. The dramatic increase of videos demands more efficient and accurate access to video content.…”
Abstract. Automatic face recognition has long been established as one of the most active research areas in computer vision. Face recognition in unconstrained environments remains challenging for most practical applications. In contrast to traditional still-image based approaches, recently the research focus has shifted towards videobased approaches. Video data provides rich and redundant information, which can be exploited to resolve the inherent ambiguities of image-based recognition like sensitivity to low resolution, pose variations and occlusion, leading to more accurate and robust recognition. Face recognition has also been considered in the content-based video retrieval setup, for example, character-based video search. In this chapter, we review existing research on face recognition and retrieval in video. The relevant techniques are comprehensively surveyed and discussed.
“…Face recognition could be used for video content description, indexing, and retrieval [25,96]. The dramatic increase of videos demands more efficient and accurate access to video content.…”
Abstract. Automatic face recognition has long been established as one of the most active research areas in computer vision. Face recognition in unconstrained environments remains challenging for most practical applications. In contrast to traditional still-image based approaches, recently the research focus has shifted towards videobased approaches. Video data provides rich and redundant information, which can be exploited to resolve the inherent ambiguities of image-based recognition like sensitivity to low resolution, pose variations and occlusion, leading to more accurate and robust recognition. Face recognition has also been considered in the content-based video retrieval setup, for example, character-based video search. In this chapter, we review existing research on face recognition and retrieval in video. The relevant techniques are comprehensively surveyed and discussed.
“…Markov Models and a k-means clustering algorithm ( [12]). The resulting clusters are supposed to represent the di erent individuals shown in the video.…”
Section: Human Being Detection Based Indexingmentioning
confidence: 99%
“…In [12], [...] frames of a video sequence are scanned for faces by a Neural Networkbased face detector. The extracted faces are then grouped into clusters by a combination of a face recognition method using pseudo two-dimensional Hidden…”
Section: Human Being Detection Based Indexingmentioning
This work proposes a human interaction recognition based approach to video indexing that represents a video by showing when and with whom was interacted throughout the video. In order to visualize the length of an interaction, it is required to recognize individuals that have been detected in earlier parts of the video. To solve this problem, an approach to photo-clustering is extended to video material by tracking detected faces and using the information from tracking to improve the recognition of human beings. The results of the tracking based approach show a considerable decrease of false cluster assignments compared to the original method. Further, it is demonstrated that the proposed method is able to correctly recognize the appearance of ve out of the six individuals correctly.iii iv
“…In [20], a spectral clustering algorithm, called the Ng algorithm [1], is used to cluster a relatively small set of faces (39 faces for 6 peoples) described by their eigenface projection. In [21], faces are described by a complex 2D-HMM model and the resulting features are simply clustered using the k-means algorithm. In [7], an affine invariant affinity measure is proposed to describe each face followed by a hierarchical k-medoids clustering technique.…”
Section: Introductionmentioning
confidence: 99%
“…While automatic face recognition aims at matching observed faces against a face database [8] [9][12] [17] [21], automatic casting consists in forming groups of similar faces represented by keyfaces (i.e. principal actors) [12].…”
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.