WideDepth: Millimeter-Accurate Benchmark for Fisheye Depth Estimation
Quick Take
WideDepth introduces the first indoor dataset for fisheye depth estimation, featuring 101 scenes with 5K stereo pairs labeled with millimeter-level ground truth. The dataset enhances performance of pinhole-trained stereo models by up to 62% on fisheye data, advancing robotics perception.
Key Points
- WideDepth features 101 scenes with 5K high-resolution stereo pairs.
- Dataset provides millimeter-level ground truth depth and disparity.
- Includes paired pinhole and fisheye samples across various setups.
- Achieves up to 62% performance boost for pinhole-based stereo models.
- Evaluates state-of-the-art monocular depth and stereo matching models.
Article Content
From source RSS / original summaryarXiv:2605. 24074v1 Announce Type: new Abstract: Fisheye cameras are increasingly adopted in robotics for near-field manipulation, navigation, and immersive perception, yet indoor depth benchmarks with accurate ground truth are still missing. To address this, we introduce WideDepth - the first indoor dataset for fisheye depth estimation, featuring 101 scenes containing 5K high-resolution stereo pairs labeled with millimeter-level ground truth depth and disparity.
Our dataset also includes paired pinhole and fisheye samples across varying fields of view and baselines in both horizontal and vertical stereo setups. We further propose a method to adapt pinhole-trained stereo models to fisheye images and introduce a novel stereo fisheye image generation pipeline based on high-resolution LiDAR scans. Leveraging these methods, we thoroughly evaluate state-of-the-art monocular depth, stereo matching, and depth completion models on our benchmark.
Additionally, we provide 18K LiDAR-derived sparse depth training samples, achieving up to a 62% performance boost on fisheye data when fine-tuning pinhole-based stereo models. In summary, the high precision and versatility of our benchmark set a strong foundation for advancing research in fisheye depth estimation and robotics perception. Project page: https://ilyaind. github. io/WideDepth
Reader Mode unavailable (could not extract clean content).
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CV
See more →Evi-Steer: Learning to Steer Biomedical Vision-Language Models through Efficient and Generalizable Evidential Tuning
Evi-Steer introduces a novel evidential tuning framework for BiomedCLIP, achieving 0.11% parameter updates while enhancing uncertainty-aware fine-tuning. It outperforms state-of-the-art methods across 15 biomedical imaging datasets, proving effective in few-shot learning and domain shifts for clinical applications.
