Towards a proactive biologically-inspired Advanced Driver Assistance System

A biologically-inspired vision architecture for resource-constrained intelligent vehicles

“…Also measuring and mimicking the reactions of an experienced driver is envisioned in the future. Our system extensions introduced in [42,45] contains first approaches towards such an efficient cognitive control concept. The central assumption is that a robust learning system requires a generic system structure with a high number of degrees of freedom for controlling the system reaction and measuring the system state.…”

“…In a more recent version of our ADAS these weights are computed dynamically at runtime (see [42]). In our current (see [45]) and future work it is envisioned to use attention weights to track and even learn new objects.…”

Towards a closer fusion of active and passive safety: Optical flow-based detection of vehicle side collisions

“…Therefore, a BEV warping approach is proposed that compensates the ego-motion. The BEV differential image is post-filtered using biologically motivated filters for denoising and contour extraction (refer to [15]). The detection results are used for estimating further object parameters.…”

Non-intrusive car driver's emotion recognition using thermal camera

“…In order to improve traffic safety researchers focused on developing ADAS applications as listed in the references [7][8][9]. Several driver monitoring systems (fatigue/emotions) have been proposed in the past which include intrusive and nonintrusive techniques [10][11][12][13][14][15][16][17][18][19][20][21].…”