Object detection with single camera stereo

Abstract

Many fielded mobile robot systems have demonstrated the importance of directly estimating the 3D shape of objects in the robot's vicinity. The most mature solutions available today use active laser scanning or stereo camera pairs, but both approaches require specialized and expensive sensors. In prior publications, we have demonstrated the generation of stereo images from a single very low-cost camera using structure from motion (SFM) techniques. In this paper we demonstrate the practical usage of single-camera stereo in real-world mobile robot applications. Stereo imagery tends to produce incomplete 3D shape reconstructions of man-made objects because of smooth/glary regions that defeat stereo matching algorithms. We demonstrate robust object detection despite such incompleteness through matching of simple parameterized geometric models. Results are presented where parked cars are detected, and then recognized via license plate recognition, all in real time by a robot traveling through a parking lot.

Single Camera Object Detection for Self-Driving Vehicle: A Review

“…According to McBride et al (2006), the weakness from a stereo image of a low-cost camera can be settled down by matching its geometrical models. In his detection over a parking lot scene, the detection rate recorded was 81.5% where it successfully detected 106 vehicles out of 130.…”

“…Jeon et al (2016) used a single stereo camera of 640x360 pixel resolutions named VSTC-V260 with 24fps in the speed average of 40kmph in testing his pedestrian detection, traffic light and traffic sign recognition. According to McBride et al (2006), the object detection can also perform well in the ease of a parking lot scene just by using a low-cost single-stereo camera. Haloi and Jayagopi (2015) used a wide angle camera sensor mounted in the vehicle's roof in capturing broad road environments which can give up to 440x680 images size at the speed of 45km/h.…”