Generalization Evaluation of Deep Stereo Matching Methods for UAV-Based Forestry Applications
Yida Lin, Bing Xue, Mengjie Zhang, Sam Schofield, Richard Green
TL;DR
The paper addresses the challenge of cross-domain generalization in deep stereo matching for UAV forestry. It conducts a comprehensive zero-shot evaluation of eight methods spanning iterative refinement, foundation models, and zero-shot adaptation across ETH3D, KITTI, Middlebury, and a new Canterbury forestry dataset, revealing that foundation models excel on structured scenes while iterative methods offer robust cross-domain performance; it also highlights dataset-specific failures like RAFT-Stereo on ETH3D. DEFOM emerges as the gold-standard baseline for vegetation depth due to its depth smoothness and occlusion handling, enabling reliable depth reference generation for forestry scenes. The work advocates multi-benchmark validation and provides deployment guidance across indoor, automotive, and safety-critical UAV applications, along with plans for public Canterbury benchmarking and uncertainty quantification.
Abstract
Autonomous UAV forestry operations require robust depth estimation methods with strong cross-domain generalization. However, existing evaluations focus on urban and indoor scenarios, leaving a critical gap for specialized vegetation-dense environments. We present the first systematic zero-shot evaluation of eight state-of-the-art stereo methods--RAFT-Stereo, IGEV, IGEV++, BridgeDepth, StereoAnywhere, DEFOM (plus baseline methods ACVNet, PSMNet, TCstereo)--spanning iterative refinement, foundation model, and zero-shot adaptation paradigms. All methods are trained exclusively on Scene Flow and evaluated without fine-tuning on four standard benchmarks (ETH3D, KITTI 2012/2015, Middlebury) plus a novel 5,313-pair Canterbury forestry dataset captured with ZED Mini camera (1920x1080). Performance reveals scene-dependent patterns: foundation models excel on structured scenes (BridgeDepth: 0.23 px on ETH3D, 0.83-1.07 px on KITTI; DEFOM: 0.35-4.65 px across benchmarks), while iterative methods maintain cross-domain robustness (IGEV++: 0.36-6.77 px; IGEV: 0.33-21.91 px). Critical finding: RAFT-Stereo exhibits catastrophic ETH3D failure (26.23 px EPE, 98 percent error rate) due to negative disparity predictions, while performing normally on KITTI (0.90-1.11 px). Qualitative evaluation on Canterbury forestry dataset identifies DEFOM as the optimal gold-standard baseline for vegetation depth estimation, exhibiting superior depth smoothness, occlusion handling, and cross-domain consistency compared to IGEV++, despite IGEV++'s finer detail preservation.