# Exploring intermediate representation for monocular vehicle pose estimation

arXiv: https://arxiv.org/abs/2011.08464 · HTML: https://mlsanity.com/paper/2011.08464/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 17 Nov 2020.

## Built on by (most cited first, 12 of 12)

- [3D Object Detection from Images for Autonomous Driving: A Survey](https://mlsanity.com/paper/2202.02980/) (2022)
- [YOLOPose: Transformer-based Multi-Object 6D Pose Estimation using Keypoint Regression](https://mlsanity.com/paper/2205.02536/) (2022)
- [CheckerPose: Progressive Dense Keypoint Localization for Object Pose Estimation with Graph Neural Network](https://mlsanity.com/paper/2303.16874/) (2023)
- [MonoEdge: Monocular 3D Object Detection Using Local Perspectives](https://mlsanity.com/paper/2301.01802/) (2023)
- [Stereo Neural Vernier Caliper](https://mlsanity.com/paper/2203.11018/) (2022)
- [YOLOPose V2: Understanding and Improving Transformer-based 6D Pose Estimation](https://mlsanity.com/paper/2307.11550/) (2023)
- [Rotation Matters: Generalized Monocular 3D Object Detection for Various Camera Systems](https://mlsanity.com/paper/2310.05366/) (2023)
- [Joint stereo 3D object detection and implicit surface reconstruction](https://mlsanity.com/paper/2111.12924/) (2021)
- [Attention is All They Need: Exploring the Media Archaeology of the Computer Vision Research Paper](https://mlsanity.com/paper/2209.11200/) (2022)
- [L3: Accelerator-Friendly Lossless Image Format for High-Resolution, High-Throughput DNN Training](https://mlsanity.com/paper/2208.08711/) (2022)
- [Object-Attribute-Relation Representation Based Video Semantic Communication](https://mlsanity.com/paper/2406.10469/) (2024)
- [Hierarchical Feature-level Reverse Propagation for Post-Training Neural Networks](https://mlsanity.com/paper/2506.07188/) (2025)

## Abstract (arXiv)

We present a new learning-based framework to recover vehicle pose in SO(3) from a single RGB image. In contrast to previous works that map from local appearance to observation angles, we explore a progressive approach by extracting meaningful Intermediate Geometrical Representations (IGRs) to estimate egocentric vehicle orientation. This approach features a deep model that transforms perceived intensities to IGRs, which are mapped to a 3D representation encoding object orientation in the camera coordinate system. Core problems are what IGRs to use and how to learn them more effectively. We answer the former question by designing IGRs based on an interpolated cuboid that derives from primitive 3D annotation readily. The latter question motivates us to incorporate geometry knowledge with a new loss function based on a projective invariant. This loss function allows unlabeled data to be used in the training stage to improve representation learning. Without additional labels, our system outperforms previous monocular RGB-based methods for joint vehicle detection and pose estimation on the KITTI benchmark, achieving performance even comparable to stereo methods. Code and pre-trained models are available at this https URL.