An efficient method for face retrieval from large video datasets

Nguyen, Thao; Ngo, Thanh Duc; Le, Duy-Dinh; Satoh, Shin’ichi; Le, Bac; Duong, Duc Anh

doi:10.1145/1816041.1816096

Cited by 13 publications

(11 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As shown in Table 5, K-Faces doesn't perform well in terms of mAP, although it had better performance for the news videos used in Ref. [3]. The reason is that the six types of videos in our dataset have larger variations than news videos, which causes the method to fail when the K faces of two face-tracks have different poses, illumination conditions and so forth.…”

Section: Comparing With Different Individual Retrieval Approachesmentioning

confidence: 93%

“…We compare our retrieval scheme to two kinds of individual retrieval approaches: i) K-Faces [3], an approach of focusing on reducing the number of feature vectors, where K faces were selected from the face-track for matching; ii) clubs [6], an approach of focusing on narrowing the scope of retrieval, which used a structure of clubs to introduce a query into one or several clusters. The mAPs and retrieval times are given in Table 5.…”

Section: Comparing With Different Individual Retrieval Approachesmentioning

confidence: 99%

“…In Ref. [3], K faces selected from each face-track were used to compute the similarity. Although the face-track based approaches showed better performance than those using single faces, they will still need a long time to do matching because a large number of face-tracks can be generated from a largescale video dataset due to the sensitivity of most face trackers on the changes of illumination, occlusions, false face detection, and so on.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Efficient Large-Scale Video Retrieval via Discriminative Signatures

Hao

Kamata

2013

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

Pengyi HAO †a) , Student Member and Sei-ichiro KAMATA †b) , Member SUMMARYThe topic of retrieving videos containing a desired person from a dataset just using the content of faces without any help of textual information has many interesting applications like video surveillance, social network, video mining, etc. However, traditional face matching against a huge number of detected faces leads to an unacceptable response time and may also reduce the accuracy due to the large variations in facial expressions, poses, lighting, etc. Therefore, in this paper we propose a novel method to generate discriminative "signatures" for efficiently retrieving the videos containing the same person with a query. In this research, the signature is defined as a compact, discriminative and reduced dimensionality representation, which is generated from a set of high-dimensional feature vectors of an individual. The desired videos are retrieved based on the similarities between the signature of the query and those of individuals in the database. In particular, we make the following contributions. Firstly, we give an algorithm of two directional linear discriminant analysis with maximum correntropy criterion (2DLDA-MCC) as an extension to our recently proposed maximum correntropy criterion based linear discriminant analysis (LDA-MCC). Both algorithms are robust to outliers and noise. Secondly, we present an approach for transferring a set of exemplars to a fixed-length signature using LDA-MCC and 2DLDA-MCC, resulting in two kinds of signatures that are called 1D signature and 2D signature. Finally, a novel video retrieval scheme is given based on the signatures, which has low storage requirement and can achieve a fast search. Evaluations on a large dataset of videos show reliable measurement of similarities by using the proposed signatures to represent the identities generated from videos. Experimental results also demonstrate that the proposed video retrieval scheme has the potential to substantially reduce the response time and slightly increase the mean average precision of retrieval. key words : individual retrieval, signature, linear discriminant analysis, video dataset

show abstract

Section: Comparing With Different Individual Retrieval Approachesmentioning

confidence: 93%

Section: Comparing With Different Individual Retrieval Approachesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Efficient Large-Scale Video Retrieval via Discriminative Signatures

Hao

Kamata

2013

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

show abstract

“…Another approach for dealing with face retrieval from a large video dataset was proposed in Ref. [5], where the mean face from a subset of faces selected from the original face track (this method was called 'k-Faces' in this study) was used to compute the similarity with k faces selected from the face tracks in the dataset. Although good performance was exhibited for news videos in Ref.…”

Section: Introductionmentioning

confidence: 99%

“…Although good performance was exhibited for news videos in Ref. [5], if a huge number of face tracks are extracted, matching between face tracks will be ineffective. In addition, in the above studies, no results were given for the formation of face tracks from different types of videos.…”

Section: Introductionmentioning

confidence: 99%

Efficiently Finding Individuals from Video Dataset

Hao

Kamata

2012

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

Pengyi HAO†a) , Nonmember and Sei-ichiro KAMATA †b) , Member SUMMARYWe are interested in retrieving video shots or videos containing particular people from a video dataset. Owing to the large variations in pose, illumination conditions, occlusions, hairstyles and facial expressions, face tracks have recently been researched in the fields of face recognition, face retrieval and name labeling from videos. However, when the number of face tracks is very large, conventional methods, which match all or some pairs of faces in face tracks, will not be effective. Therefore, in this paper, an efficient method for finding a given person from a video dataset is presented. In our study, in according to performing research on face tracks in a single video, we also consider how to organize all the faces in videos in a dataset and how to improve the search quality in the query process. Different videos may include the same person; thus, the management of individuals in different videos will be useful for their retrieval. The proposed method includes the following three points. (i) Face tracks of the same person appearing for a period in each video are first connected on the basis of scene information with a time constriction, then all the people in one video are organized by a proposed hierarchical clustering method. (ii) After obtaining the organizational structure of all the people in one video, the people are organized into an upper layer by affinity propagation. (iii) Finally, in the process of querying, a remeasuring method based on the index structure of videos is performed to improve the retrieval accuracy. We also build a video dataset that contains six types of videos: films, TV shows, educational videos, interviews, press conferences and domestic activities. The formation of face tracks in the six types of videos is first researched, then experiments are performed on this video dataset containing more than 1 million faces and 218,786 face tracks. The results show that the proposed approach has high search quality and a short search time.

show abstract