[code]004⼈脸识别数据集之
YouTubeFaceLabeledFacesinthe。。。
快与慢作文1) Face DataSet:海参淡干
1. CASIA-3D FaceV1宁德白水洋
collected a 3D face databa consisting of 4624 scans of 123 persons using the non-contact 3D digitizer, Minolta Vivid 910, as shown in Fig.1. During building the databa, we consider not only the single variations of pos, expressions and illuminations, but also the combined variations of expressions under illumination and pos under expressions, as shown in Fig.2, Fig.3 and Fig.4. To the subjects with glass, we will collect one additional scans with glass. Thus, each person contains 37 or 38 scans. And from each scan, one 2D color image and one 3D facial triangulated surface are also generated. We aims to build a complete 3D face databa, which is further driven to be a public platform in testing the algorithms in 3D face recognition or others.
2. Casia-FaceV5
包含500个⼈的照⽚，每个⼈5张，共2500张照⽚。照⽚size：height 480,width 640。
3 The CNBC Face Databa
This face databa was created by the Tarrlab at Brown University (Tarrlab is now at Carnegie Mellon).
This databa includes multiple images for over 200 individuals of many different races with consistent lighting, multiple views, real emotions, and disguis (and some participants returned for a cond ssion veral weeks later with a haircut, or a new beard, etc.).
听起来像英语
4. YouTube Face
The data t contains 3,425 videos of 1,595 different people. All the videos were downloaded from . An average of 2.15
videos are available for each subject. The shortest clip duration is 48 frames, the longest clip is 6,070 frames, and the
average length of a video clip is 181.3 frames.
宫颈炎最佳治疗方法5. 31_Person identification in TV ries
招商引资工作Person identification in TV ries(1.1GB
Face tracks, features and shot boundaries from our latest CVPR 2013 paper. It is obtained from 6 episodes of Buffy the Vamp
生日快乐贺卡
link
6. CMUVASC & PIE Face datat
包括来⾃68个⼈的40000张照⽚，其中包括了每个⼈的13种姿态条件，43种光照条件和4种表情下的照⽚，现有的多姿态⼈脸识别的⽂献
基本上都是在CMU PIE⼈脸库上测试的什么是低筋面粉
7. Labeled Faces in the Wild Home
a databa of face photographs designed for studying the problem of unconstrained face recognition. The data t
contains more than 13,000 images of faces collected from the web. Each face has been labeled with the name of the
person pictured. 1680 of the people pictured have two or more distinct photos in the data t. The on
ly constraint on the faces is that they were detected by the Viola-Jones face detector. More details can be found in the technical report below
8. BioID Face Databa - FaceDB
The datat consists of 1521 gray level images with a resolution of 384x286 pixel. Each one shows the frontal view of a face
9. Multi-Task Facial Landmark (MTFL) datat
Facial landmark detection of face alignment has long been impeded by the problems of occlusion and po variation. Instead of treating the detection task as a single and independent problem, we investigate the possibility of improving detection robustness through multi-task learning. Specifically, we wish to optimize facial landmark detection together with heterogeneous but subtly correlated tasks, e.g., head po estimation and facial attribute inference. This is non-trivial since different tasks have different learning difficulties and convergence rates. To address this problem, we formulate a novel tasks-constrained deep model, with task-wi early stopping to facilitate learning c
onvergence. Extensive evaluations show that the propod task-constrained learning (i) outperforms existing methods, especially in dealing with faces with vere occlusion and po variation, and (ii) reduces model complexity drastically compared to the state-of-the-art method bad on cascaded deep model.
10. colorferet
