Name
Abstract | ||
|---|---|---|
Face detectors are usually trained on static images but deployed in the wild such as surveillance videos. Due to the domain shift between images and videos, directly applying the image-based face detectors onto videos usually gives unsatisfactory performance. In this paper, we introduce the BoxFlow - a new unsupervised detector adaptation method that can effectively adapt a face detector pre-trained on static images to videos. BoxFlow unsupervisedly adapts face detectors through fully exploiting the motion contexts across video frames. In particular, BoxFlow introduces three novel components: (1) generalized heat map representation of face locations with augmented shape flexibility; (2) motion based temporal contextual regularization among adjacent frames for unsupervised face detection refinement; (3) a self-paced learning strategy that adapts face detectors from easy data samples to challenging ones progressively. With these key components, we develop a systematic unsupervised face detector adaptation framework to help face detectors adapt to various deployed environments. Extensive experiments on the IDA dataset clearly demonstrate the superiority of our proposed method. Without utilizing any annotation, the BoxFlow achieves about 10%-20% performance gain in terms of Average Precision than directly applying image-based face detectors. |
Year | DOI | Venue |
|---|---|---|
2017 | 10.1109/FG.2017.46 | 2017 12th IEEE International Conference on Automatic Face & Gesture Recognition (FG 2017) |
Keywords | Field | DocType |
BoxFlow,unsupervised face detector adaptation,static images,surveillance videos,image-based face detectors,face detector pretraining,motion contexts,video frames,generalized heat map representation,face locations,augmented shape flexibility,motion based temporal contextual regularization,unsupervised face detection refinement,self-paced learning strategy,IDA dataset,average precision | Computer vision,Annotation,Object-class detection,Computer science,Regularization (mathematics),Artificial intelligence,Face detection,Detector | Conference |
ISSN | ISBN | Citations |
2326-5396 | 978-1-5090-4024-7 | 0 |
PageRank | References | Authors |
0.34 | 19 | 4 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Jianshu Li | 1 | 141 | 12.04 |
| Jiashi Feng | 2 | 2165 | 140.81 |
| Luoqi Liu | 3 | 397 | 18.64 |
| Terence Sim | 4 | 2562 | 169.42 |