Geometry-constrained Car Recognition Using a 3D Perspective Network

03/19/2019
by   Rui Zeng, et al.
0

We present a novel learning framework for vehicle recognition from a single RGB image. Unlike existing methods which only use attention mechanisms to locate 2D discriminative information, our unified framework learns a joint representation of the 2D global texture and 3D-bounding-box in a mutually correlated and reinforced way. These two kinds of feature representation are combined by a novel fusion network, which predicts the vehicle's category. The 2D global feature is extracted using an off-the-shelf detection network, where the estimated 2D bounding box assists in finding the region of interest (RoI). With the assistance of the RoI, the 3D bounding box and its corresponding features are generated in a geometrically correct way using a novel 3D perspective Network (3DPN). The 3DPN consists of a convolutional neural network (CNN), a vanishing point loss, and RoI perspective layers. The CNN regresses the 3D bounding box under the guidance of the proposed vanishing point loss, which provides a perspective geometry constraint. Thanks to the proposed RoI perspective layer, the variation caused by viewpoint changes is corrected via the estimated geometry, enhancing the feature representation. We present qualitative and quantitative results for our approach on the vehicle classification and verification tasks in the BoxCars dataset. The results demonstrate that, by learning how to extract features from the 3D bounding box, we can achieve comparable or superior performance to methods that only use 2D information.

READ FULL TEXT

page 1

page 5

page 7

research
03/19/2019

3DCarRecog: Car Recognition Using 3D Bounding Box

We present a novel learning framework for vehicle recognition from a sin...
research
01/27/2019

6D Object Pose Estimation Based on 2D Bounding Box

In this paper, we present a simple but powerful method to tackle the pro...
research
03/26/2019

GS3D: An Efficient 3D Object Detection Framework for Autonomous Driving

We present an efficient 3D object detection framework based on a single ...
research
02/09/2020

Weakly Supervised Attention Pyramid Convolutional Neural Network for Fine-Grained Visual Classification

Classifying the sub-categories of an object from the same super-category...
research
11/29/2017

PointFusion: Deep Sensor Fusion for 3D Bounding Box Estimation

We present PointFusion, a generic 3D object detection method that levera...
research
10/20/2014

Supervised mid-level features for word image representation

This paper addresses the problem of learning word image representations:...
research
10/22/2017

Deep Cropping via Attention Box Prediction and Aesthetics Assessment

We model the photo cropping problem as a cascade of attention box regres...

Please sign up or login with your details

Forgot password? Click here to reset