Name
Abstract | ||
|---|---|---|
Cascades help make sliding window object detection fast, nevertheless, computational demands remain prohibitive for numerous applications. Currently, evaluation of adjacent windows proceeds independently; this is suboptimal as detector responses at nearby locations and scales are correlated. We propose to exploit these correlations by tightly coupling detector evaluation of nearby windows. We introduce two opposing mechanisms: detector excitation of promising neighbors and inhibition of inferior neighbors. By enabling neighboring detectors to communicate, crosstalk cascades achieve major gains (4-30× speedup) over cascades evaluated independently at each image location. Combined with recent advances in fast multi-scale feature computation, for which we provide an optimized implementation, our approach runs at 35-65 fps on 640×480 images while attaining state-of-the-art accuracy. |
Year | DOI | Venue |
|---|---|---|
2012 | 10.1007/978-3-642-33709-3_46 | ECCV (2) |
Keywords | Field | DocType |
nearby windows,neighboring detector,computational demand,adjacent windows proceed,detector excitation,crosstalk cascade,fast multi-scale feature computation,cascades help,nearby location,frame-rate pedestrian detection,detector response,coupling detector evaluation | Object detection,Sliding window protocol,Coupling,Computer science,Real-time computing,Frame rate,Pedestrian detection,Detector,Speedup,Computation | Conference |
Volume | ISSN | Citations |
7573 | 0302-9743 | 105 |
PageRank | References | Authors |
4.68 | 21 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Piotr Dollár | 1 | 7999 | 307.07 |
| Ron Appel | 2 | 405 | 14.78 |
| Wolf Kienzle | 3 | 391 | 20.73 |