# Are we Missing Confidence in Pseudo-LiDAR Methods for Monocular 3D Object Detection?

arXiv: https://arxiv.org/abs/2012.05796 · HTML: https://mlsanity.com/paper/2012.05796/

## Key facts

- Cited by 27 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 10 Dec 2020.

## Built on by (most cited first, 27 of 27)

- [Vision-Centric BEV Perception: A Survey](https://mlsanity.com/paper/2208.02797/) (2022)
- [3D Object Detection from Images for Autonomous Driving: A Survey](https://mlsanity.com/paper/2202.02980/) (2022)
- [Augmenting Lane Perception and Topology Understanding with Standard Definition Navigation Maps](https://mlsanity.com/paper/2311.04079/) (2023)
- [DEVIANT: Depth EquiVarIAnt NeTwork for Monocular 3D Object Detection](https://mlsanity.com/paper/2207.10758/) (2022)
- [Diversity Matters: Fully Exploiting Depth Clues for Reliable Monocular 3D Object Detection](https://mlsanity.com/paper/2205.09373/) (2022)
- [Localization Is All You Evaluate: Data Leakage in Online Mapping Datasets and How to Fix It](https://mlsanity.com/paper/2312.06420/) (2023)
- [Monocular 3D Object Detection with Depth from Motion](https://mlsanity.com/paper/2207.12988/) (2022)
- [Pseudo-Stereo for Monocular 3D Object Detection in Autonomous Driving](https://mlsanity.com/paper/2203.02112/) (2022)
- [MonoJSG: Joint Semantic and Geometric Cost Volume for Monocular 3D Object Detection](https://mlsanity.com/paper/2203.08563/) (2022)
- [Lidar Point Cloud Guided Monocular 3D Object Detection](https://mlsanity.com/paper/2104.09035/) (2021)
- [LaRa: Latents and Rays for Multi-Camera Bird's-Eye-View Semantic Segmentation](https://mlsanity.com/paper/2206.13294/) (2022)
- [Mix-Teaching: A Simple, Unified and Effective Semi-Supervised Learning Framework for Monocular 3D Object Detection](https://mlsanity.com/paper/2207.04448/) (2022)
- [PointBeV: A Sparse Approach to BeV Predictions](https://mlsanity.com/paper/2312.00703/) (2023)
- [NeurOCS: Neural NOCS Supervision for Monocular 3D Object Localization](https://mlsanity.com/paper/2305.17763/) (2023)
- [RICCARDO: Radar Hit Prediction and Convolution for Camera-Radar 3D Object Detection](https://mlsanity.com/paper/2504.09086/) (2025)
- [Monocular 3D Object Detection with Bounding Box Denoising in 3D by Perceiver](https://mlsanity.com/paper/2304.01289/) (2023)
- [Boosting Monocular 3D Object Detection with Object-Centric Auxiliary Depth Supervision](https://mlsanity.com/paper/2210.16574/) (2022)
- [Attention-Based Depth Distillation with 3D-Aware Positional Encoding for Monocular 3D Object Detection](https://mlsanity.com/paper/2211.16779/) (2022)
- [Shape-Aware Monocular 3D Object Detection](https://mlsanity.com/paper/2204.08717/) (2022)
- [GUPNet++: Geometry Uncertainty Propagation Network for Monocular 3D Object Detection](https://mlsanity.com/paper/2310.15624/) (2023)
- [Camera Height Doesn't Change: Unsupervised Training for Metric Monocular Road-Scene Depth Estimation](https://mlsanity.com/paper/2312.04530/) (2023)
- [An Empirical Study of Pseudo-Labeling for Image-based 3D Object Detection](https://mlsanity.com/paper/2208.07137/) (2022)
- [Towards Fair and Comprehensive Comparisons for Image-Based 3D Object Detection](https://mlsanity.com/paper/2310.05447/) (2023)
- [OriCon3D: Effective 3D Object Detection using Orientation and Confidence](https://mlsanity.com/paper/2304.14484/) (2023)
- [SVDM: Single-View Diffusion Model for Pseudo-Stereo 3D Object Detection](https://mlsanity.com/paper/2307.02270/) (2023)
- [Efficient Feature Aggregation and Scale-Aware Regression for Monocular 3D Object Detection](https://mlsanity.com/paper/2411.02747/) (2024)
- [A Data-Centric Approach to 3D Semantic Segmentation of Railway Scenes](https://mlsanity.com/paper/2504.18213/) (2025)

## Abstract (arXiv)

Pseudo-LiDAR-based methods for monocular 3D object detection have received considerable attention in the community due to the performance gains exhibited on the KITTI3D benchmark, in particular on the commonly reported validation split. This generated a distorted impression about the superiority of Pseudo-LiDAR-based (PL-based) approaches over methods working with RGB images only. Our first contribution consists in rectifying this view by pointing out and showing experimentally that the validation results published by PL-based methods are substantially biased. The source of the bias resides in an overlap between the KITTI3D object detection validation set and the training/validation sets used to train depth predictors feeding PL-based methods. Surprisingly, the bias remains also after geographically removing the overlap. This leaves the test set as the only reliable set for comparison, where published PL-based methods do not excel. Our second contribution brings PL-based methods back up in the ranking with the design of a novel deep architecture which introduces a 3D confidence prediction module. We show that 3D confidence estimation techniques derived from RGB-only 3D detection approaches can be successfully integrated into our framework and, more importantly, that improved performance can be obtained with a newly designed 3D confidence measure, leading to state-of-the-art performance on the KITTI3D benchmark.