FastViDAR: Real-Time Omnidirectional Depth Estimation via Alternative Hierarchical Attention
Not on the map. Why?
Abstract (arXiv)
In this paper we propose FastViDAR, a novel framework that takes four fisheye camera inputs and produces a full 360^circ depth map along with per-camera depth, fusion depth, and confidence estimates. Our main contributions are: (1) We introduce Alternative Hierarchical Attention (AHA) mechanism that efficiently fuses features across views through separate intra-frame and inter-frame windowed self-attention, achieving cross-view feature mixing with reduced overhead. (2) We propose a novel ERP fusion approach that projects multi-view depth estimates to a shared equirectangular coordinate system to obtain the final fusion depth. (3) We generate ERP image-depth pairs using HM3D and 2D3D-S datasets for comprehensive evaluation, demonstrating competitive zero-shot performance on real datasets while achieving up to 20 FPS on NVIDIA Orin NX embedded hardware. Project page: hrefhttps://3f7dfc.github.io/FastVidar/https://3f7dfc.github.io/FastVidar/
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 22 of its 40 references that are in the map. First posted on arXiv 28 Sep 2025.
Builds on
22 of 40 references.
References · 22
- 2017Attention Is All You Needcited by 60,361
- 20233D Gaussian Splatting for Real-Time Radiance Field Renderingcited by 6,226
- 2017CARLA: An Open Urban Driving Simulatorcited by 2,449
- 2025VGGT: Visual Geometry Grounded Transformercited by 1,539
- 2021CvT: Introducing Convolutions to Vision Transformerscited by 947
- 2015A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimationcited by 881
- 2018MVSNet: Depth Inference for Unstructured Multi-view Stereocited by 788
- 2021Habitat-Matterport 3D Dataset (HM3D): 1000 Large-scale 3D Environments for Embodied AIcited by 592
- 2017Joint 2D-3D-Semantic Data for Indoor Scene Understandingcited by 406
- 2020Virtual KITTI 2cited by 353
- 2019Unsupervised Scale-consistent Depth and Ego-motion Learning from Monocular Videocited by 244
- 2020D3VO: Deep Depth, Deep Pose and Deep Uncertainty for Monocular Visual Odometrycited by 206
10 more
- 2022Practical Stereo Matching via Cascaded Recurrent Network with Adaptive Correlationcited by 197
- 2025FoundationStereo: Zero-Shot Stereo Matchingcited by 173
- 2018Falling Things: A Synthetic Dataset for 3D Object Detection and Pose Estimationcited by 92
- 2023FasterViT: Fast Vision Transformers with Hierarchical Attentioncited by 65
- 2018The Double Sphere Camera Modelcited by 52
- 2022Open Challenges in Deep Stereo: the Booster Datasetcited by 48
- 2025Depth Any Camera: Zero-Shot Metric Depth Estimation from Any Cameracited by 40
- 2024LightStereo: Channel Boost Is All You Need for Efficient 2D Cost Aggregationcited by 28
- 2019SweepNet: Wide-baseline Omnidirectional Depth Estimationcited by 24
- 2022FreDSNet: Joint Monocular Depth and Semantic Segmentation with Fast Fourier Convolutionscited by 11
Built on by
Citing papers per year: 2026: 2.
2 papers.