A One-and-Half Stage Pedestrian Detector
| Autor: | Francois Bremond, Aziz Dziri, Bertrand Leroy, Ujjwal Ujjwal |
|---|---|
| Přispěvatelé: | Spatio-Temporal Activity Recognition Systems (STARS), Inria Sophia Antipolis - Méditerranée (CRISAM), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), COMUE Université Côte d'Azur (2015-2019) (COMUE UCA), VEhicule DEcarboné et COmmuniquant et sa Mobilité (VeDeCom), Ujjwal, Ujjwal |
| Jazyk: | angličtina |
| Rok vydání: | 2020 |
| Předmět: |
[INFO.INFO-AI] Computer Science [cs]/Artificial Intelligence [cs.AI]
Computer science business.industry Pedestrian detection Pooling Detector Inference 020207 software engineering 02 engineering and technology Object detection [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI] Feature (computer vision) 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Segmentation Computer vision Artificial intelligence business Selection (genetic algorithm) |
| Zdroj: | WACV 2020-IEEE Winter Conference on Applications of Computer Vision WACV 2020-IEEE Winter Conference on Applications of Computer Vision, Mar 2020, Snowmass Village, United States WACV |
| Popis: | International audience; Pedestrian detection is a specific instance of the more general problem of object detection in computer vision. A balance between detection accuracy and speed is a desirable trait for pedestrian detection systems in many applications such as self-driving cars. In this paper, we follow the wisdom of " and less is often more" to achieve this balance. We propose a lightweight mechanism based on semantic segmentation to reduce the number of anchors to be processed. We furthermore unify this selection with the intra-anchor feature pooling strategy adopted in high performance two-stage detectors such as Faster-RCNN. Such a strategy is avoided in one-stage detectors like SSD in favour of faster inference but at the cost of reducing the accuracy vis-à-vis two-stage detectors. However our anchor selection renders it practical to use feature pooling without giving up the inference speed. Our proposed approach succeeds in detecting pedestrians with state-of-art performance on caltech-reasonable and ciypersons datasets with inference speeds of ∼ 32 fps. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|