CudaSIFT-SLAM：用于真实人体内窥镜全流程建图的多地图视觉SLAM
CudaSIFT-SLAM: multiple-map visual SLAM for full procedure mapping in real human endoscopy
AI总结：
针对人体结肠镜中视觉SLAM频繁跟踪丢失的问题，提出CudaSIFT-SLAM，采用SIFT特征和暴力匹配替代ORB和DBoW2，实现实时多地图建图与重定位，显著提升建图覆盖率。
摘要：
单目视觉同步定位与建图（V-SLAM）如今在移动机器人和增强现实中是不可或缺的工具，并且在这些领域表现稳健。然而，人体结肠镜检查带来了严峻的挑战，如遮挡、模糊、光照变化、缺乏纹理、变形、水流喷射或工具交互，这些都会导致非常频繁的跟踪丢失。ORB-SLAM3作为性能最优的多地图V-SLAM系统，由于其基于ORB特征和DBoW2词袋的场所识别算法性能不佳，无法通过合并子地图或重定位相机来从这些情况中恢复。我们提出了CudaSIFT-SLAM，这是第一个能够实时处理完整人体结肠镜检查的V-SLAM系统。为了克服ORB-SLAM3的局限性，我们使用SIFT代替ORB特征，并将DBoW2直接索引替换为计算量更大的暴力匹配，从而能够成功匹配时间上分离的图像，用于重定位和地图合并。实时性能得益于CudaSIFT，这是一种用于SIFT提取和暴力匹配的GPU实现。我们在C3VD体模结肠数据集和一个来自Endomapper数据集的完整真实结肠镜检查中对我们的系统进行了基准测试，展示了合并子地图并在其中重定位的能力，获得了显著更长的子地图。我们的系统在C3VD数据集中成功实时建图覆盖了88%的帧。在真实筛查结肠镜检查中，尽管遮挡和模糊帧的比例更高，但在仔细探索区域的建图覆盖率为53%，在完整序列中为38%，比ORB-SLAM3提高了70%。
英文摘要：
Monocular visual simultaneous localization and mapping (V-SLAM) is nowadays an irreplaceable tool in mobile robotics and augmented reality, where it performs robustly. However, human colonoscopies pose formidable challenges like occlusions, blur, light changes, lack of texture, deformation, water jets or tool interaction, which result in very frequent tracking losses. ORB-SLAM3, the top performing multiple-map V-SLAM, is unable to recover from them by merging sub-maps or relocalizing the camera, due to the poor performance of its place recognition algorithm based on ORB features and DBoW2 bag-of-words. We present CudaSIFT-SLAM, the first V-SLAM system able to process complete human colonoscopies in real-time. To overcome the limitations of ORB-SLAM3, we use SIFT instead of ORB features and replace the DBoW2 direct index with the more computationally demanding brute-force matching, being able to successfully match images separated in time for relocation and map merging. Real-time performance is achieved thanks to CudaSIFT, a GPU implementation for SIFT extraction and brute-force matching. We benchmark our system in the C3VD phantom colon dataset, and in a full real colonoscopy from the Endomapper dataset, demonstrating the capabilities to merge sub-maps and relocate in them, obtaining significantly longer sub-maps. Our system successfully maps in real-time 88 % of the frames in the C3VD dataset. In a real screening colonoscopy, despite the much higher prevalence of occluded and blurred frames, the mapping coverage is 53 % in carefully explored areas and 38 % in the full sequence, a 70 % improvement over ORB-SLAM3.
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。
