{"id":11272,"date":"2026-08-21T07:23:50","date_gmt":"2026-08-20T23:23:50","guid":{"rendered":"https:\/\/infernews.com\/blog\/official-implementation-of-the-paper-quot-towards-real-time-and-adaptable-lidar\/"},"modified":"2026-08-21T07:25:45","modified_gmt":"2026-08-20T23:25:45","slug":"official-implementation-of-the-paper-quot-towards-real-time-and-adaptable-lidar","status":"publish","type":"post","link":"https:\/\/infernews.com\/blog\/official-implementation-of-the-paper-quot-towards-real-time-and-adaptable-lidar\/","title":{"rendered":"RapidLiDAR\uff1a\u55ae\u6b21\u524d\u5411\u50b3\u64ad\u505a\u5230 10 Hz \u5605 LiDAR \u5834\u666f\u88dc\u5168"},"content":{"rendered":"\n
\"Repository<\/figure>\n\n\n\n

\u5c0d\u505a\u81ea\u52d5\u99d5\u99db\u6216\u6a5f\u68b0\u4eba\u611f\u77e5\u5605\u4eba\u569f\u8b1b\uff0cLiDAR \u5834\u666f\u88dc\u5168\u6700\u982d\u75db\u5605\u5514\u4fc2\u8cea\u7d20\uff0c\u800c\u4fc2\u6162\u2014\u2014\u597d\u591a\u73fe\u6709\u65b9\u6cd5\u8981\u5e7e\u767e\u6beb\u79d2\u5148\u51fa\u4e00\u5e40\uff0c\u6839\u672c\u8ffd\u5514\u4e0a\u8eca\u898f\u6216\u5373\u6642\u53cd\u61c9\u5605\u9700\u6c42\u3002RapidLiDAR \u5605\u5b9a\u4f4d\u5c31\u55ba\u5462\u5ea6\uff1a\u4f62\u8a2d\u8a08\u6210\u4e00\u500b\u7aef\u5230\u7aef\u3001\u55ae\u6b21\u524d\u5411\u50b3\u64ad\u5605\u88dc\u5168\u6a21\u578b\uff0c\u76ee\u6a19\u4fc2\u8dd1\u5230\u63a5\u8fd1 10 Hz\uff0c\u540c\u6642\u4fdd\u7559\u53ef\u8abf\u901f\u5ea6\u540c\u53ef\u8abf\u7bc0\u8f38\u51fa\u5bc6\u5ea6\u5605\u5f48\u6027\u3002<\/p>\n\n\n\n

\u505a\u6cd5\u4e0a\uff0c\u4f62\u5148\u7528\u9ad4\u7d20\u5316\u62bd\u53d6\u591a\u5c3a\u5ea6 3D \u7279\u5fb5\uff0c\u518d\u900f\u904e\u4e00\u500b\u81ea\u6ce8\u610f\u529b BEV \u982d\u7522\u751f\u5bc6\u96c6 2D \u7279\u5fb5\u5716\u3002Adaptive Initialization Module \u6703\u9810\u6e2c\u4e00\u500b\u7a7a\u9593\u8b8a\u5316\u5605\u4f4d\u79fb\uff0c\u628a\u7a00\u758f\u9ede\u96f2\u300c\u6490\u958b\u300d\u505a\u7c97\u7565\u521d\u59cb\u5316\uff1b\u4e4b\u5f8c Multi-Scale Reconstruction Module \u7528\u53ef\u8b8a\u5f62\u6ce8\u610f\u529b\uff08deformable attention\uff09\u5c07\u6bcf\u9ede\u7279\u5fb5\u540c\u591a\u5c3a\u5ea6 BEV \u7279\u5fb5\u5c0d\u9f4a\u505a\u7cbe\u4fee\u3002\u5982\u679c\u60f3\u518d\u7a20\u5bc6\u4ef2\u53ef\u4ee5\u52a0\u4e00\u500b\u53ef\u9078\u5605 Refinement Network \u55ba\u51cd\u7d50\u5605\u4e3b\u6a21\u578b\u4e0a\u505a\u4e0a\u63a1\u6a23\uff0c\u500d\u7387\u70ba \u03ba\u3002<\/p>\n\n\n\n

\u63db\u53e5\u8a71\u8b1b\uff0c\u820a\u65b9\u6cd5\u901a\u5e38\u5206\u958b\u505a\u521d\u59cb\u5316\u540c\u8fed\u4ee3 refinement\uff0c\u6216\u8005\u7528\u5169\u968e\u6bb5\u7db2\u7d61\u5148\u7c97\u5f8c\u7d30\uff1bRapidLiDAR \u5c07\u5462\u5e7e\u6b65\u58d3\u5165\u4e00\u6b21\u524d\u5411\uff0c\u9760 BEV \u5ea7\u6a19\u5316\u540c\u53ef\u8b8a\u5f62\u6ce8\u610f\u529b\u540c\u6642\u517c\u9867\u6548\u7387\u540c\u5e7e\u4f55\u4e00\u81f4\u6027\u3002\u4ee3\u50f9\u4fc2\u4f9d\u8cf4 SemanticKITTI \u98a8\u683c\u5605\u8cc7\u6599 loader\uff0c\u9700\u8981 .npy<\/code> \u683c\u5f0f\u5605 GT \u540c input \u914d\u5c0d\uff0c\u5514\u4fc2 out-of-the-box \u8655\u7406\u539f\u59cb Velodyne \u6383\u63cf\u3002<\/p>\n\n\n\n

\u9069\u5408\u505a\u81ea\u52d5\u99d5\u99db stack\u3001\u5be6\u6642 SLAM\u3001\u6a5f\u68b0\u4eba\u611f\u77e5\u539f\u578b\uff0c\u6216\u8005\u60f3\u55ba\u908a\u7de3\u88dd\u7f6e\u4e0a\u8a66\u7a20\u5bc6 LiDAR \u9810\u6e2c\u5605\u5718\u968a\u3002\u539f\u6587\u5f37\u8abf\u5373\u6642\u6027\u70ba 10 Hz\uff0c\u5177\u9ad4 mIoU \u6216 Chamfer distance \u7b49\u6578\u5b57\u672a\u5728 README \u5b8c\u6574\u5217\u51fa\uff0c\u5b89\u88dd\u9700 CUDA 12.4 \u540c PyTorch 2.4.1\uff0c\u4e26\u8981\u81ea\u884c JIT \u7de8\u8b6f Chamfer distance CUDA extension\u3002<\/p>\n\n\n\n

GitHub<\/strong><\/a> \u00b7 Paper<\/strong><\/a><\/p>\n","protected":false},"excerpt":{"rendered":"

\u4f62\u5c07\u7a00\u758f\u5605\u90e8\u5206\u9ede\u96f2\u4e00\u6b21\u6027\u9084\u539f\u6210\u7a20\u5bc6\u5834\u666f\uff0c\u4e3b\u6253\u5373\u6642\u540c\u53ef\u8abf\u901f\u5ea6\uff0c\u9069\u5408\u5c0d\u5ef6\u9072\u654f\u611f\u5605\u81ea\u52d5\u99d5\u99db\u540c\u6a5f\u5668\u4eba\u958b\u767c\u8005\u3002<\/p>\n","protected":false},"author":8,"featured_media":11271,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"wpai_generated_summary":"","footnotes":""},"categories":[133,179,132,152,76,149,184,199],"tags":[],"class_list":["post-11272","post","type-post","status-publish","format-standard","hentry","category-133","category-nvidia","category-3d","category-python","category-76","category-149","category-robotic","category-dataset-"],"_links":{"self":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/posts\/11272","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/users\/8"}],"replies":[{"embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/comments?post=11272"}],"version-history":[{"count":1,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/posts\/11272\/revisions"}],"predecessor-version":[{"id":11274,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/posts\/11272\/revisions\/11274"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/media\/11271"}],"wp:attachment":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/media?parent=11272"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/categories?post=11272"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/tags?post=11272"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}