TY - GEN
T1 - Multimodal and multiview distillation for real-time player detection on a football field
AU - Cioppa, Anthony
AU - Deliege, Adrien
AU - Huda, Noor Ul
AU - Gade, Rikke
AU - Droogenbroeck, Marc Van
AU - Moeslund, Thomas B.
PY - 2020/6
Y1 - 2020/6
N2 - Monitoring the occupancy of public sports facilities is essential to assess their use and to motivate their construction in new places. In the case of a football field, the area to cover is large, thus several regular cameras should be used, which makes the setup expensive and complex. As an alternative, we developed a system that detects players from a unique cheap and wide-angle fisheye camera assisted by a single narrow-angle thermal camera. In this work, we train a network in a knowledge distillation approach in which the student and the teacher have different modalities and a different view of the same scene. In particular, we design a custom data augmentation combined with a motion detection algorithm to handle the training in the region of the fisheye camera not covered by the thermal one. We show that our solution is effective in detecting players on the whole field filmed by the fisheye camera. We evaluate it quantitatively and qualitatively in the case of an online distillation, where the student detects players in real time while being continuously adapted to the latest video conditions.
AB - Monitoring the occupancy of public sports facilities is essential to assess their use and to motivate their construction in new places. In the case of a football field, the area to cover is large, thus several regular cameras should be used, which makes the setup expensive and complex. As an alternative, we developed a system that detects players from a unique cheap and wide-angle fisheye camera assisted by a single narrow-angle thermal camera. In this work, we train a network in a knowledge distillation approach in which the student and the teacher have different modalities and a different view of the same scene. In particular, we design a custom data augmentation combined with a motion detection algorithm to handle the training in the region of the fisheye camera not covered by the thermal one. We show that our solution is effective in detecting players on the whole field filmed by the fisheye camera. We evaluate it quantitatively and qualitatively in the case of an online distillation, where the student detects players in real time while being continuously adapted to the latest video conditions.
UR - http://www.scopus.com/inward/record.url?scp=85090146665&partnerID=8YFLogxK
U2 - 10.1109/CVPRW50498.2020.00448
DO - 10.1109/CVPRW50498.2020.00448
M3 - Article in proceeding
T3 - IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
SP - 3846
EP - 3855
BT - IEEE Conference on Computer Vision and Pattern Recognition Workshops
PB - IEEE
T2 - 2020 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
Y2 - 14 June 2020 through 19 June 2020
ER -