Learning And-Or Models to Represent Context and Occlusion for Car Detection and Viewpoint Estimation

01/29/2015
by   Tianfu Wu, et al.
0

This paper presents a method for learning And-Or models to represent context and occlusion for car detection and viewpoint estimation. The learned And-Or model represents car-to-car context and occlusion configurations at three levels: (i) spatially-aligned cars, (ii) single car under different occlusion configurations, and (iii) a small number of parts. The And-Or model embeds a grammar for representing large structural and appearance variations in a reconfigurable hierarchy. The learning process consists of two stages in a weakly supervised way (i.e., only bounding boxes of single cars are annotated). Firstly, the structure of the And-Or model is learned with three components: (a) mining multi-car contextual patterns based on layouts of annotated single car bounding boxes, (b) mining occlusion configurations between single cars, and (c) learning different combinations of part visibility based on car 3D CAD simulation. The And-Or model is organized in a directed and acyclic graph which can be inferred by Dynamic Programming. Secondly, the model parameters (for appearance, deformation and bias) are jointly trained using Weak-Label Structural SVM. In experiments, we test our model on four car detection datasets --- the KITTI dataset Geiger12, the PASCAL VOC2007 car dataset pascal, and two self-collected car datasets, namely the Street-Parking car dataset and the Parking-Lot car dataset, and three datasets for car viewpoint estimation --- the PASCAL VOC2006 car dataset pascal, the 3D car dataset savarese, and the PASCAL3D+ car dataset xiang_wacv14. Compared with state-of-the-art variants of deformable part-based models and other methods, our model achieves significant improvement consistently on the four detection datasets, and comparable performance on car viewpoint estimation.

READ FULL TEXT

page 1

page 2

page 3

page 7

page 8

page 11

page 12

page 14

research
06/09/2014

Parsing Semantic Parts of Cars Using Graphical Models and Segment Appearance Consistency

This paper addresses the problem of semantic part parsing (segmentation)...
research
03/26/2016

Recognizing Car Fluents from Video

Physical fluents, a term originally used by Newton [40], refers to time-...
research
12/25/2014

Joint Deep Learning for Car Detection

Traditional object recognition approaches apply feature extraction, part...
research
10/08/2020

A Comparative Study on Effects of Original and Pseudo Labels for Weakly Supervised Learning for Car Localization Problem

In this study, the effects of different class labels created as a result...
research
11/27/2018

Part-level Car Parsing and Reconstruction from Single Street View

In this paper, we make the first attempt to build a framework to simulta...
research
08/31/2020

Analysis and Prediction of Deforming 3D Shapes using Oriented Bounding Boxes and LSTM Autoencoders

For sequences of complex 3D shapes in time we present a general approach...
research
04/10/2019

Google Street View image of a house predicts car accident risk of its resident

Road traffic injuries are a leading cause of death worldwide. Proper est...

Please sign up or login with your details

Forgot password? Click here to reset