3D Object Class Detection in the Wild
This addresses the problem of precise 3D scene understanding for computer vision applications, representing an incremental advance by building on existing 2D detection and 2D-3D lifting techniques.
The paper tackles 3D object class detection by designing a method that enriches 2D detection outputs with viewpoint, keypoints, and 3D shape estimates, achieving state-of-the-art performance in simultaneous 2D bounding box and viewpoint estimation on the Pascal3D+ dataset.
Object class detection has been a synonym for 2D bounding box localization for the longest time, fueled by the success of powerful statistical learning techniques, combined with robust image representations. Only recently, there has been a growing interest in revisiting the promise of computer vision from the early days: to precisely delineate the contents of a visual scene, object by object, in 3D. In this paper, we draw from recent advances in object detection and 2D-3D object lifting in order to design an object class detector that is particularly tailored towards 3D object class detection. Our 3D object class detection method consists of several stages gradually enriching the object detection output with object viewpoint, keypoints and 3D shape estimates. Following careful design, in each stage it constantly improves the performance and achieves state-ofthe-art performance in simultaneous 2D bounding box and viewpoint estimation on the challenging Pascal3D+ dataset.