Mono3D++: Monocular 3D Vehicle Detection with Two-Scale 3D Hypotheses and Task Priors

2019/01/11 by Tong He, Stefano Soatto, He, Tong +1 · 2 citations
Computer Science · Engineering · #Advanced Neural Network Applications #Advanced Vision and Imaging #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Robotics and Sensor-Based Localization

paper · pdf · doi:10.48550/arxiv.1901.03446

openalex publication_date 2019/01/11 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We present a method to infer 3D pose and shape of vehicles from a single image. To tackle this ill-posed problem, we optimize two-scale projection consistency between the generated 3D hypotheses and their 2D pseudo-measurements. Specifically, we use a morphable wireframe model to generate a fine-scaled representation of vehicle shape and pose. To reduce its sensitivity to 2D landmarks, we jointly model the 3D bounding box as a coarse representation which improves robustness. We also integrate three task priors, including unsupervised monocular depth, a ground plane constraint as well as vehicle shape priors, with forward projection errors into an overall energy function.

Citations

• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• Shape Manifolds, Procrustean Metrics, and Complex Projective Spaces
• Stacked Hourglass Networks for Human Pose Estimation
• PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space

Cited by

• Few-Shot Learning in Video and 3D Object Detection: A Survey
• ToosiCubix: Monocular 3D Cuboid Labeling via Vehicle Part Annotations

Related

• Vehicle Pose and Shape Estimation through Multiple Monocular Vision
• Consistency of Implicit and Explicit Features Matters for Monocular 3D Object Detection
• RTM3D: Real-time Monocular 3D Detection from Object Keypoints for Autonomous Driving
• What My Motion tells me about Your Pose: A Self-Supervised Monocular 3D\n Vehicle Detector
• Monocular 3D Vehicle Detection Using Uncalibrated Traffic Cameras through Homography