ForestVO：通过 ForestGlue 增强森林环境中的视觉里程计
ForestVO: Enhancing Visual Odometry in Forest Environments through ForestGlue
AI总结：
本文提出 ForestGlue 与 ForestVO 方法，通过优化特征检测与匹配并引入基于 transformer 的位姿估计，构建了专为森林环境设计的端到端视觉里程计流水线，在大幅降低计算开销的同时显著提升了动态场景下的导航精度与鲁棒性。
摘要：
视觉里程计系统的最新进展改善了自主导航；然而，在森林等复杂环境中仍存在挑战，密集的树叶、多变的光照和重复的纹理会损害特征匹配的准确性。为了应对这些挑战，我们引入了 ForestGlue，通过四种配置——灰度、RGB、RGB-D 和立体视觉——增强 SuperPoint 特征检测器，针对各种传感模态进行了优化。在特征匹配方面，我们使用 LightGlue 或 SuperGlue，并用合成森林数据重新训练。ForestGlue 实现了与基线模型相当的位姿估计精度，但仅需 512 个关键点——仅为基线 2048 个关键点的 25%——即可在 10° 阈值下达到 0.745 的 LO-RANSAC AUC 分数。由于仅需四分之一的关键点，ForestGlue 显著降低了计算开销，证明了其在动态森林环境中的有效性，并使其适合在资源受限的平台上进行实时部署。通过将 ForestGlue 与基于 transformer 的位姿估计模型相结合，我们提出了 ForestVO，它利用帧间匹配的 2D 像素坐标来估计相对相机位姿。在具有挑战性的 TartanAir 森林序列上，ForestVO 实现了 1.09 m 的平均相对位姿误差（RPE）和 2.33% 的 kitti_score，在动态场景中比 DSO 等直接法高出 40%。尽管仅使用 10% 的数据集进行训练，ForestVO 仍保持了与 TartanVO 相当的竞争性能，同时是一个显著更轻量的模型。这项工作建立了一个专为森林环境中的视觉里程计量身定制的端到端深度学习流水线，利用特定于森林的训练数据来优化特征匹配和位姿估计，从而增强了自主导航系统的准确性和鲁棒性。
英文摘要：
Recent advancements in visual odometry systems have improved autonomous navigation; however, challenges persist in complex environments like forests, where dense foliage, variable lighting, and repetitive textures compromise feature correspondence accuracy. To address these challenges, we introduce ForestGlue, enhancing the SuperPoint feature detector through four configurations - grayscale, RGB, RGB-D, and stereo-vision - optimised for various sensing modalities. For feature matching, we employ LightGlue or SuperGlue, retrained with synthetic forest data. ForestGlue achieves comparable pose estimation accuracy to baseline models but requires only 512 keypoints - just 25% of the baseline's 2048 - to reach an LO-RANSAC AUC score of 0.745 at a 10° threshold. With only a quarter of keypoints needed, ForestGlue significantly reduces computational overhead, demonstrating effectiveness in dynamic forest environments, and making it suitable for real-time deployment on resource-constrained platforms. By combining ForestGlue with a transformer-based pose estimation model, we propose ForestVO, which estimates relative camera poses using matched 2D pixel coordinates between frames. On challenging TartanAir forest sequences, ForestVO achieves an average relative pose error (RPE) of 1.09 m and a kitti_score of 2.33%, outperforming direct-based methods like DSO by 40% in dynamic scenes. Despite using only 10% of the dataset for training, ForestVO maintains competitive performance with TartanVO while being a significantly lighter model. This work establishes an end-to-end deep learning pipeline specifically tailored for visual odometry in forested environments, leveraging forest-specific training data to optimise feature correspondence and pose estimation, thereby enhancing the accuracy and robustness of autonomous navigation systems.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
