# Toward Real-world BEV Perception: Depth Uncertainty Estimation via Gaussian Splatting

arXiv: https://arxiv.org/abs/2504.01957 · HTML: https://mlsanity.com/paper/2504.01957/

## Key facts

- Cited by 20 papers in the map (data through 14 Oct 2026).
- Builds on 30 of its 39 references that are in the map.
- First posted on arXiv 2 Apr 2025.

## Builds on (most cited first, 30 of 30)

- [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://mlsanity.com/paper/1905.11946/) (2019)
- [3D Gaussian Splatting for Real-Time Radiance Field Rendering](https://mlsanity.com/paper/2308.04079/) (2023)
- [nuScenes: A multimodal dataset for autonomous driving](https://mlsanity.com/paper/1903.11027/) (2019)
- [BEVFormer: Learning Bird's-Eye-View Representation from Multi-Camera Images via Spatiotemporal Transformers](https://mlsanity.com/paper/2203.17270/) (2022)
- [Lift, Splat, Shoot: Encoding Images From Arbitrary Camera Rigs by Implicitly Unprojecting to 3D](https://mlsanity.com/paper/2008.05711/) (2020)
- [BEVFusion: Multi-Task Multi-Sensor Fusion with Unified Bird's-Eye View Representation](https://mlsanity.com/paper/2205.13542/) (2022)
- [BEVDet: High-performance Multi-camera 3D Object Detection in Bird-Eye-View](https://mlsanity.com/paper/2112.11790/) (2021)
- [ZoeDepth: Zero-shot Transfer by Combining Relative and Metric Depth](https://mlsanity.com/paper/2302.12288/) (2023)
- [BEVDepth: Acquisition of Reliable Depth for Multi-view 3D Object Detection](https://mlsanity.com/paper/2206.10092/) (2022)
- [PETR: Position Embedding Transformation for Multi-View 3D Object Detection](https://mlsanity.com/paper/2203.05625/) (2022)
- [ST-P3: End-to-end Vision-based Autonomous Driving via Spatial-Temporal Feature Learning](https://mlsanity.com/paper/2207.07601/) (2022)
- [PETRv2: A Unified Framework for 3D Perception from Multi-Camera Images](https://mlsanity.com/paper/2206.01256/) (2022)
- [Cross-view Transformers for real-time Map-view Semantic Segmentation](https://mlsanity.com/paper/2205.02833/) (2022)
- [Deep Deterministic Uncertainty: A Simple Baseline](https://mlsanity.com/paper/2102.11582/) (2021)
- [On the uncertainty of self-supervised monocular depth estimation](https://mlsanity.com/paper/2005.06209/) (2020)
- [M2BEV: Multi-Camera Joint 3D Detection and Segmentation with Unified Birds-Eye View Representation](https://mlsanity.com/paper/2204.05088/) (2022)
- [Simple-BEV: What Really Matters for Multi-Sensor BEV Perception?](https://mlsanity.com/paper/2206.07959/) (2022)
- [ViP3D: End-to-end Visual Trajectory Prediction via 3D Agent Queries](https://mlsanity.com/paper/2208.01582/) (2022)
- [FB-BEV: BEV Representation from Forward-Backward View Transformations](https://mlsanity.com/paper/2308.02236/) (2023)
- [BEVStereo: Enhancing Depth Estimation in Multi-view 3D Object Detection with Dynamic Temporal Stereo](https://mlsanity.com/paper/2209.10248/) (2022)
- [NeRF On-the-go: Exploiting Uncertainty for Distractor-free NeRFs in the Wild](https://mlsanity.com/paper/2405.18715/) (2024)
- [IEBins: Iterative Elastic Bins for Monocular Depth Estimation](https://mlsanity.com/paper/2309.14137/) (2023)
- [Efficient Uncertainty Estimation for Semantic Segmentation in Videos](https://mlsanity.com/paper/1807.11037/) (2018)
- [Producing and Leveraging Online Map Uncertainty in Trajectory Prediction](https://mlsanity.com/paper/2403.16439/) (2024)
- [LaRa: Latents and Rays for Multi-Camera Bird's-Eye-View Semantic Segmentation](https://mlsanity.com/paper/2206.13294/) (2022)
- [Gradient-based Uncertainty for Monocular Depth Estimation](https://mlsanity.com/paper/2208.02005/) (2022)
- [Neural Visibility Field for Uncertainty-Driven Active Mapping](https://mlsanity.com/paper/2406.06948/) (2024)
- [PointBeV: A Sparse Approach to BeV Predictions](https://mlsanity.com/paper/2312.00703/) (2023)
- [Estimating 3D Uncertainty Field: Quantifying Uncertainty for Neural Radiance Fields](https://mlsanity.com/paper/2311.01815/) (2023)
- [Quantifying Uncertainty in Motion Prediction with Variational Bayesian Mixture](https://mlsanity.com/paper/2404.03789/) (2024)

## Built on by (most cited first, 20 of 20)

- [SQS: Enhancing Sparse Perception Models via Query-based Splatting in Autonomous Driving](https://mlsanity.com/paper/2509.16588/) (2025)
- [JointSplat: Probabilistic Joint Flow-Depth Optimization for Sparse-View Gaussian Splatting](https://mlsanity.com/paper/2506.03872/) (2025)
- [RadarGaussianDet3D: Gaussian Representation-based Real-time 3D Object Detection with 4D Automotive Radars](https://mlsanity.com/paper/2509.16119/) (2025)
- [RCTDistill: Cross-Modal Knowledge Distillation Framework for Radar-Camera 3D Object Detection with Temporal Fusion](https://mlsanity.com/paper/2509.17712/) (2025)
- [UniMPR: A Unified Framework for Multimodal Place Recognition with Heterogeneous Sensor Configurations](https://mlsanity.com/paper/2512.18279/) (2025)
- [Benchmarking PhD-Level Coding in 3D Geometric Computer Vision](https://mlsanity.com/paper/2603.30038/) (2026)
- [DLWM: Dual Latent World Models enable Holistic Gaussian-centric Pre-training in Autonomous Driving](https://mlsanity.com/paper/2604.00969/) (2026)
- [Progressive Bird's Eye View Perception for Safety-Critical Autonomous Driving: A Comprehensive Survey](https://mlsanity.com/paper/2508.07560/) (2025)
- [FisheyeGaussianLift: BEV Feature Lifting for Surround-View Fisheye Camera Perception](https://mlsanity.com/paper/2511.17210/) (2025)
- [SUPER-AD: Semantic Uncertainty-aware Planning for End-to-End Robust Autonomous Driving](https://mlsanity.com/paper/2511.22865/) (2025)
- [Reconstruction Matters: Learning Geometry-Aligned BEV Representation through 3D Gaussian Splatting](https://mlsanity.com/paper/2603.19193/) (2026)
- [Towards 3D Object-Centric Feature Learning for Semantic Scene Completion](https://mlsanity.com/paper/2511.13031/) (2025)
- [HyGE-Occ: Hybrid View-Transformation with 3D Gaussian and Edge Priors for 3D Panoptic Occupancy Prediction](https://mlsanity.com/paper/2512.19871/) (2025)
- [Fusion4CA: Boosting 3D Object Detection via Comprehensive Image Exploitation](https://mlsanity.com/paper/2603.05305/) (2026)
- [RCGDet3D: Rethinking 4D Radar-Camera Fusion-based 3D Object Detection with Enhanced Radar Feature Encoding](https://mlsanity.com/paper/2605.21112/) (2026)
- [Open-Vocabulary BEV Segmentation with 3D-Aware Geometric Constraints](https://mlsanity.com/paper/2606.24353/) (2026)
- [GaussianMap: Learning Gaussian Representation for Multi-Sensor Online HD Map Construction](https://mlsanity.com/paper/2606.31177/) (2026)
- [FlashBEV: Fast and Memory-Efficient Exact BEV Transformation with IO-Awareness](https://mlsanity.com/paper/2607.10071/) (2026)
- [RayOcc: Occlusion-Aware Ray Occupancy Estimation via Gaussian Mixture Intensity](https://mlsanity.com/paper/2607.17660/) (2026)
- [EgoNeMo: Transferable Map of Pedestrian Dynamics via Egocentric LiDAR Scan](https://mlsanity.com/paper/2609.06195/) (2026)

## Abstract (arXiv)

Bird's-eye view (BEV) perception has gained significant attention because it provides a unified representation to fuse multiple view images and enables a wide range of down-stream autonomous driving tasks, such as forecasting and planning. Recent state-of-the-art models utilize projection-based methods which formulate BEV perception as query learning to bypass explicit depth estimation. While we observe promising advancements in this paradigm, they still fall short of real-world applications because of the lack of uncertainty modeling and expensive computational requirement. In this work, we introduce GaussianLSS, a novel uncertainty-aware BEV perception framework that revisits unprojection-based methods, specifically the Lift-Splat-Shoot (LSS) paradigm, and enhances them with depth un-certainty modeling. GaussianLSS represents spatial dispersion by learning a soft depth mean and computing the variance of the depth distribution, which implicitly captures object extents. We then transform the depth distribution into 3D Gaussians and rasterize them to construct uncertainty-aware BEV features. We evaluate GaussianLSS on the nuScenes dataset, achieving state-of-the-art performance compared to unprojection-based methods. In particular, it provides significant advantages in speed, running 2.5x faster, and in memory efficiency, using 0.3x less memory compared to projection-based methods, while achieving competitive performance with only a 0.4% IoU difference.