Using an Intelligent Vision System for Obstacle Detection in Winter Condition

Ziadia, Marwa; Kélouwani, Sousso; Amamou, Ali; Dubé, Yves; Agbossou, Kodjo

doi:10.5220/0007766900002179

Cited by 1 publication

(2 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It supports the dramatic impact of winter on object detection. Ziadia et al [10] also observed that winter conditions reduce pedestrian and distant vehicles' detection rate. Chebrolu et al [11] have mentioned that most models fail to predict pedestrians in a dark environment.…”

Section: Introductionmentioning

confidence: 96%

See 1 more Smart Citation

Attention transfer from human to neural networks for road object detection in winter

Boisclair

Kélouwani

Ayevide

et al. 2022

IET Image Processing

Self Cite

View full text Add to dashboard Cite

As an essential feature of autonomous road vehicles, obstacle detection must be executed on a real‐time onboard platform with high accuracy. Cameras are still the most commonly used sensors in autonomous driving. Most detections using cameras are based on convolutional neural networks. In this regard, a recent teacher–student approach, called transfer learning, has been used to improve the neural network training process. This approach has only been used with a neural network acting as a teacher to the best of our knowledge. This paper proposes a novel way of improving training data based on attention transfer by getting the attention map from a human. The proposed method allows the dataset size reduction by 50%, which leads to up to a 60% decline in the training time. The experimental results indicate that the proposed method can enhance the F1‐score of the network by up to 10% in winter conditions.

show abstract

Section: Introductionmentioning

confidence: 96%

“…Ziadia et al. [10] also observed that winter conditions reduce pedestrian and distant vehicles' detection rate. Chebrolu et al.…”

Section: Introductionmentioning

confidence: 99%