{"id":21945,"date":"2024-08-29T13:22:26","date_gmt":"2024-08-29T05:22:26","guid":{"rendered":"https:\/\/aif.amtbbs.org\/?p=21945"},"modified":"2024-08-29T13:22:26","modified_gmt":"2024-08-29T05:22:26","slug":"%e5%85%a5%e8%a1%8c%e8%87%aa%e5%8a%a8%e9%a9%be%e9%a9%b6%e6%95%b0%e6%8d%ae%e9%97%ad%e7%8e%af%ef%bc%8c%e4%bb%8a%e5%b9%b4%e5%bf%85%e8%af%bb%e7%9a%84%e5%8d%81%e4%b8%89%e7%af%87%e6%9c%80%e5%89%8d%e6%b2%bf","status":"publish","type":"post","link":"https:\/\/aif.amtbbs.org\/index.php\/2024\/08\/29\/21945\/","title":{"rendered":"\u5165\u884c\u81ea\u52a8\u9a7e\u9a76\u6570\u636e\u95ed\u73af\uff0c\u4eca\u5e74\u5fc5\u8bfb\u7684\u5341\u4e09\u7bc7\u6700\u524d\u6cbf\u8bba\u6587"},"content":{"rendered":"
\"\"<\/div>\n
<\/div>\n
\u4eca\u5929\u4e3a\u5927\u5bb6\u76d8\u70b93DGS&NeRF\u5728\u81ea\u52a8\u9a7e\u9a76\u91cd\u5efa\u65b9\u9762\u7684\u5de5\u4f5c\uff01<\/div>\n
\n
\n

\u672c\u6587\u7ecf\u81ea\u52a8\u9a7e\u9a76\u4e4b\u5fc3\u516c\u4f17\u53f7\u6388\u6743\u8f6c\u8f7d\uff0c\u8f6c\u8f7d\u8bf7\u8054\u7cfb\u51fa\u5904\u3002<\/p>\n

\u8fd1\u51e0\u5e74\uff0c\u81ea\u52a8\u9a7e\u9a76\u6280\u672f\u7684\u53d1\u5c55\u65e5\u65b0\u6708\u5f02\u3002\u4eceECCV 2020\u7684NeRF\u95ee\u4e16\u518d\u5230SIGGRAPH 2023\u76843DGS\uff0c\u4e09\u7ef4\u91cd\u5efa\u8d70\u4e0a\u4e86\u5feb\u901f\u53d1\u5c55\u7684\u9053\u8def\uff01\u518d\u5230\u81ea\u52a8\u9a7e\u9a76\u7aef\u5230\u7aef\u6280\u672f\u7684\u95ee\u4e16\uff0c\u4e0e\u4e4b\u76f8\u5173\u7684\u4eff\u771f\u95ed\u73af\u5f00\u59cb\u9891\u7e41\u51fa\u73b0\u5728\u5927\u4f17\u89c6\u91ce\u4e2d\uff0c\u65b0\u5174\u7684\u4e09\u7ef4\u91cd\u5efa\u6280\u672f\u7531\u6b64\u5728\u81ea\u52a8\u9a7e\u9a76\u9886\u57df\u4e5f\u9010\u6e10\u7115\u53d1\u65b0\u673a\u30022023\u5e748\u6708\u7279\u65af\u62c9\u53d1\u5e03FSD V12\uff1b2024\u5e744\u6708\u5546\u6c64\u7edd\u5f71\u53d1\u5e03\u9762\u5411\u91cf\u4ea7\u7684\u7aef\u5230\u7aef\u81ea\u52a8\u9a7e\u9a76\u89e3\u51b3\u65b9\u6cd5UniAD\uff1b2024\u5e747\u6708\u7406\u60f3\u590f\u5b63\u53d1\u5e03\u4f1a\u5ba3\u79f0\u7aef\u5230\u7aef\u6b63\u5f0f\u4e0a\u8f66\uff0c\u5feb\u7cfb\u7edf4D One Model\u3001\u6162\u7cfb\u7edfVLM\uff0c\u5e76\u9996\u6b21\u63d0\u51fa\u300e\u91cd\u5efa+\u751f\u6210\u300f\u7684\u4e16\u754c\u6a21\u578b\u6d4b\u8bd5\u65b9\u6848<\/strong>\u3002<\/p>\n

\u53ef\u4ee5\u8bf4\uff0c\u7aef\u5230\u7aef+\u4eff\u771f\u95ed\u73af\u662f\u5f53\u4e0b\u81ea\u52a8\u9a7e\u9a76\u53d1\u5c55\u7684\u4e3b\u6d41\u8def\u7ebf\u3002\u4f46\u662f\u4eff\u771f\u95ed\u73af\u63d0\u4e86\u5f88\u591a\u5e74\uff0c\u5230\u5e95\u4ec0\u4e48\u662f\u4eff\u771f\u95ed\u73af\uff1f\u4eff\u771f\u95ed\u73af\u7684\u6838\u5fc3\u53c8\u662f\u4ec0\u4e48\uff1f\u4e09\u7ef4\u91cd\u5efa\u53c8\u5728\u95ed\u73af\u4e2d\u8d77\u5230\u4ec0\u4e48\u6837\u7684\u4f5c\u7528\uff1f\u4e1a\u5185\u4e5f\u4e00\u76f4\u5728\u8ba8\u8bba\uff0c\u767e\u82b1\u9f50\u653e\u3002\u65e0\u8bba\u5982\u4f55\uff0c\u95ed\u73af\u7684\u76ee\u7684\u662f\u660e\u786e\u7684\uff0c\u964d\u4f4e\u5b9e\u8f66\u6d4b\u8bd5\u7684\u6210\u672c\u548c\u98ce\u9669\u3001\u6709\u6548\u63d0\u9ad8\u6a21\u578b\u7684\u5f00\u53d1\u6548\u7387\u8fdb\u800c\u4f18\u5316\u7cfb\u7edf\u6027\u80fd\u3001\u6d4b\u8bd5\u5404\u79cdcorner case\u5e76\u4f18\u5316\u6574\u4e2a\u7aef\u5230\u7aef\u7b97\u6cd5\u3002<\/p>\n

\u4eca\u5929\u5c31\u548c\u5927\u5bb6\u76d8\u4e00\u76d8\u81ea\u52a8\u9a7e\u9a76\u4e2d\u65b0\u5174\u7684\u4e09\u7ef4\u91cd\u5efa\u6280\u672f\u76f8\u5173\u7b97\u6cd5\u3002<\/p>\n

MARS: An Instance-aware, Modular and Realistic Simulator for Autonomous Driving\uff08CICAI 2023\uff09<\/strong><\/p>\n

    \n
  • \u8bba\u6587\u94fe\u63a5\uff1ahttps:\/\/arxiv.org\/abs\/2307.15058v1<\/li>\n
  • \u4ee3\u7801\u94fe\u63a5\uff1ahttps:\/\/github.com\/OPEN-AIR-SUN\/mars<\/li>\n<\/ul>\n

    \"\u56fe\u7247\"<\/p>\n

    \u6e05\u534eAIR\u63d0\u51fa\u7684\u9996\u4e2a\u5f00\u6e90\u81ea\u52a8\u9a7e\u9a76NeRF\u4eff\u771f\u5de5\u5177<\/strong>\uff01\u5982\u4eca\u81ea\u52a8\u9a7e\u9a76\u6c7d\u8f66\u5728\u666e\u901a\u60c5\u51b5\u4e0b\u53ef\u4ee5\u5e73\u7a33\u884c\u9a76\uff0c\u4eba\u4eec\u666e\u904d\u8ba4\u4e3a\uff0c\u903c\u771f\u7684\u4f20\u611f\u5668\u4eff\u771f\u5c06\u5728\u901a\u8fc7\u4eff\u771f\u89e3\u51b3\u5269\u4f59\u7684corner case\u65b9\u9762\u53d1\u6325\u5173\u952e\u4f5c\u7528\u3002\u4e3a\u6b64\uff0c\u6211\u4eec\u63d0\u51fa\u4e86\u4e00\u79cd\u57fa\u4e8e\u795e\u7ecf\u8f90\u5c04\u573a\uff08NeRFs\uff09\u7684\u81ea\u52a8\u9a7e\u9a76\u4eff\u771f\u5668\u3002\u4e0e\u73b0\u6709\u7684\u5de5\u4f5c\u76f8\u6bd4\uff0c\u6211\u4eec\u6709\u4e09\u4e2a\u663e\u8457\u7684\u7279\u70b9\uff1a<\/p>\n

      \n
    • Instance-aware\uff1a\u524d\u666f\u76ee\u6807\u548c\u80cc\u666f\uff0c\u5355\u72ec\u5efa\u6a21\uff0c\u56e0\u6b64\u53ef\u4ee5\u4fdd\u8bc1\u53ef\u63a7\u6027<\/li>\n
    • Modular\uff1a\u6a21\u5757\u5316\u8bbe\u8ba1\uff0c\u4fbf\u4e8e\u96c6\u6210\u5404\u79cdSOTA\u7684\u7b97\u6cd5\u8fdb\u6765<\/li>\n
    • Realistic\uff1a\u7531\u4e8e\u6a21\u5757\u5316\u7684\u8bbe\u8ba1\uff0c\u4e0d\u540c\u6a21\u5757\u53ef\u4ee5\u7075\u6d3b\u9009\u62e9\u6bd4\u8f83\u597d\u7684\u7b97\u6cd5\u5b9e\u73b0\uff0c\u56e0\u6b64\u6548\u679cSOTA\u3002<\/li>\n<\/ul>\n

      UniSim: A Neural Closed-Loop Sensor Simulator\uff08CVPR 2023\uff09<\/strong><\/p>\n

        \n
      • \u8bba\u6587\u94fe\u63a5\uff1ahttps:\/\/arxiv.org\/abs\/2308.01898v1<\/li>\n
      • \u9879\u76ee\u4e3b\u9875\uff1ahttps:\/\/waabi.ai\/unisim\/<\/li>\n<\/ul>\n

        \"\u56fe\u7247\"<\/p>\n

        Waabi\u548c\u591a\u4f26\u591a\u5927\u5b66\u5728CVPR 2023\u4e0a\u7684\u5de5\u4f5c\uff1a\u4e25\u683c\u6d4b\u8bd5\u81ea\u52a8\u9a7e\u9a76\u7cfb\u7edf\u5bf9\u4e8e\u5b9e\u73b0\u5b89\u5168\u7684\u81ea\u52a8\u9a7e\u9a76\u6c7d\u8f66\uff08SDV\uff09\u81f3\u5173\u91cd\u8981\u3002\u5b83\u8981\u6c42\u4eba\u4eec\u751f\u6210\u8d85\u51fa\u4e16\u754c\u4e0a\u5b89\u5168\u6536\u96c6\u8303\u56f4\u7684\u5b89\u5168\u5173\u952e\u573a\u666f\uff0c\u56e0\u4e3a\u8bb8\u591a\u573a\u666f\u5f88\u5c11\u53d1\u751f\u5728\u516c\u5171\u9053\u8def\u4e0a\u3002\u4e3a\u4e86\u51c6\u786e\u8bc4\u4f30\u6027\u80fd\uff0c\u6211\u4eec\u9700\u8981\u5728\u95ed\u73af\u4e2d\u6d4b\u8bd5\u8fd9\u4e9b\u573a\u666f\u4e2d\u7684SDV\uff0c\u5176\u4e2dSDV\u548c\u5176\u4ed6\u53c2\u4e0e\u8005\u5728\u6bcf\u4e2a\u65f6\u95f4\u6b65\u76f8\u4e92\u4f5c\u7528\u3002\u4ee5\u524d\u8bb0\u5f55\u7684\u9a7e\u9a76\u65e5\u5fd7\u4e3a\u6784\u5efa\u8fd9\u4e9b\u65b0\u573a\u666f\u63d0\u4f9b\u4e86\u4e30\u5bcc\u7684\u8d44\u6e90\uff0c\u4f46\u5bf9\u4e8e\u95ed\u73af\u8bc4\u4f30\uff0c\u6211\u4eec\u9700\u8981\u6839\u636e\u65b0\u7684\u573a\u666f\u914d\u7f6e\u548cSDV\u7684\u51b3\u5b9a\u4fee\u6539\u4f20\u611f\u5668\u6570\u636e\uff0c\u56e0\u4e3a\u53ef\u80fd\u4f1a\u6dfb\u52a0\u6216\u5220\u9664\u53c2\u4e0e\u8005\uff0c\u73b0\u6709\u53c2\u4e0e\u8005\u548cSDV\u4e4b\u95f4\u7684\u8f68\u8ff9\u5c06\u4e0e\u539f\u59cb\u8f68\u8ff9\u4e0d\u540c\u3002\u672c\u6587\u4ecb\u7ecd\u4e86UniSim\uff0c\u8fd9\u662f\u4e00\u79cd\u795e\u7ecf\u4f20\u611f\u5668\u6a21\u62df\u5668\uff0c\u5b83\u5c06\u914d\u5907\u4f20\u611f\u5668\u7684\u8f66\u8f86\u6355\u83b7\u7684\u5355\u4e2a\u8bb0\u5f55\u65e5\u5fd7\u8f6c\u6362\u4e3a\u73b0\u5b9e\u7684\u95ed\u73af\u591a\u4f20\u611f\u5668\u6a21\u62df\u3002UniSim\u6784\u5efa\u795e\u7ecf\u7279\u5f81\u7f51\u683c\u6765\u91cd\u5efa\u573a\u666f\u4e2d\u7684\u9759\u6001\u80cc\u666f\u548c\u52a8\u6001\u53c2\u4e0e\u8005\uff0c\u5e76\u5c06\u5b83\u4eec\u7ec4\u5408\u5728\u4e00\u8d77\uff0c\u4ee5\u5728\u65b0\u89c6\u89d2\u4eff\u771fLiDAR\u548c\u76f8\u673a\u6570\u636e\uff0c\u6dfb\u52a0\u6216\u5220\u9664\u53c2\u4e0e\u8005\u4ee5\u53ca\u65b0\u7684\u4f4d\u7f6e\u3002\u4e3a\u4e86\u66f4\u597d\u5730\u5904\u7406\u5916\u63a8\u89c6\u56fe\uff0c\u6211\u4eec\u4e3a\u52a8\u6001\u76ee\u6807\u5f15\u5165\u4e86\u53ef\u5b66\u4e60\u7684\u5148\u9a8c\uff0c\u5e76\u5229\u7528\u5377\u79ef\u7f51\u7edc\u6765\u5b8c\u6210\u770b\u4e0d\u89c1\u7684\u533a\u57df\u3002\u6211\u4eec\u7684\u5b9e\u9a8c\u8868\u660e\uff0cUniSim\u53ef\u4ee5\u5728\u4e0b\u6e38\u4efb\u52a1\u4e2d\u6a21\u62df\u5177\u6709\u8f83\u5c0f\u57df\u95f4\u9699\u7684\u771f\u5b9e\u4f20\u611f\u5668\u6570\u636e\u3002\u901a\u8fc7UniSim\uff0c\u6211\u4eec\u6f14\u793a\u4e86\u5728\u5b89\u5168\u5173\u952e\u573a\u666f\u4e0b\u5bf9\u81ea\u4e3b\u7cfb\u7edf\u7684\u95ed\u73af\u8bc4\u4f30\uff0c\u5c31\u50cf\u5728\u73b0\u5b9e\u4e16\u754c\u4e2d\u4e00\u6837\u3002UniSim\u7684\u4e3b\u8981\u8d21\u732e\u5982\u4e0b\uff1a<\/p>\n

          \n
        • \u9ad8\u5ea6\u903c\u771f(high realism): \u53ef\u4ee5\u51c6\u786e\u5730\u6a21\u62df\u771f\u5b9e\u4e16\u754c(\u56fe\u7247\u548cLiDAR), \u51cf\u5c0f\u9e3f\u6c9f(domain gap )<\/li>\n
        • \u95ed\u73af\u6d4b\u8bd5(closed-loop simulation): \u53ef\u4ee5\u751f\u6210\u7f55\u89c1\u7684\u5371\u9669\u573a\u666f\u6d4b\u8bd5\u65e0\u4eba\u8f66, \u5e76\u5141\u8bb8\u65e0\u4eba\u8f66\u548c\u73af\u5883\u81ea\u7531\u4ea4\u4e92<\/li>\n
        • \u53ef\u6269\u5c55 (scalable): \u53ef\u4ee5\u5f88\u5bb9\u6613\u7684\u6269\u5c55\u5230\u66f4\u591a\u7684\u573a\u666f, \u53ea\u9700\u8981\u91c7\u96c6\u4e00\u6b21\u6570\u636e, \u5c31\u80fd\u91cd\u5efa\u5e76\u4eff\u771f\u6d4b<\/li>\n
        • \u77e5\u4e4e\u89e3\u8bfb\uff1ahttps:\/\/zhuanlan.zhihu.com\/p\/636695025<\/li>\n
        • \u4e00\u4f5c\u76f4\u64ad\uff1ahttps:\/\/www.bilibili.com\/video\/BV1nj41197TZ<\/li>\n<\/ul>\n

          EmerNeRF: Emergent Spatial-Temporal Scene Decomposition via Self-Supervision<\/strong><\/p>\n

            \n
          • \u8bba\u6587\u94fe\u63a5\uff1ahttps:\/\/arxiv.org\/abs\/2311.02077v1<\/li>\n
          • \u4ee3\u7801\u94fe\u63a5\uff1ahttps:\/\/github.com\/NVlabs\/EmerNeRF<\/li>\n<\/ul>\n

            \"\u56fe\u7247\"<\/p>\n

            \u52a0\u5229\u798f\u5c3c\u4e9a\u5927\u5b66\u7684\u5de5\u4f5c\uff1a\u672c\u6587\u63d0\u51fa\u4e86EmerNeRF\uff0c\u8fd9\u662f\u4e00\u79cd\u7b80\u5355\u800c\u5f3a\u5927\u7684\u5b66\u4e60\u52a8\u6001\u9a7e\u9a76\u573a\u666f\u65f6\u7a7a\u8868\u793a\u7684\u65b9\u6cd5\u3002EmerNeRF\u4ee5\u795e\u7ecf\u573a\u4e3a\u57fa\u7840\uff0c\u901a\u8fc7\u81ea\u4e3e\u540c\u65f6\u6355\u83b7\u573a\u666f\u51e0\u4f55\u3001\u5916\u89c2\u3001\u8fd0\u52a8\u548c\u8bed\u4e49\u3002EmerNeRF\u4f9d\u8d56\u4e8e\u4e24\u4e2a\u6838\u5fc3\u7ec4\u4ef6\uff1a\u9996\u5148\uff0c\u5b83\u5c06\u573a\u666f\u5212\u5206\u4e3a\u9759\u6001\u548c\u52a8\u6001\u573a\u3002\u8fd9\u79cd\u5206\u89e3\u7eaf\u7cb9\u6e90\u4e8e\u81ea\u76d1\u7763\uff0c\u4f7f\u6211\u4eec\u7684\u6a21\u578b\u80fd\u591f\u4ece\u4e00\u822c\u7684\u3001\u91ce\u5916\u7684\u6570\u636e\u6e90\u4e2d\u5b66\u4e60\u3002\u5176\u6b21\uff0cEmerNeRF\u5c06\u52a8\u6001\u573a\u4e2d\u7684\u611f\u5e94\u6d41\u573a\u53c2\u6570\u5316\uff0c\u5e76\u4f7f\u7528\u8be5\u6d41\u573a\u8fdb\u4e00\u6b65\u805a\u5408\u591a\u5e27\u7279\u5f81\uff0c\u4ece\u800c\u63d0\u9ad8\u4e86\u52a8\u6001\u76ee\u6807\u7684\u6e32\u67d3\u7cbe\u5ea6\u3002\u8026\u5408\u8fd9\u4e09\u4e2a\u573a\uff08\u9759\u6001\u3001\u52a8\u6001\u548c\u6d41\uff09\u4f7fEmerNeRF\u80fd\u591f\u81ea\u7ed9\u81ea\u8db3\u5730\u8868\u793a\u9ad8\u5ea6\u52a8\u6001\u7684\u573a\u666f\uff0c\u800c\u65e0\u9700\u4f9d\u8d56GT\u6807\u6ce8\u6216\u9884\u5148\u8bad\u7ec3\u7684\u6a21\u578b\u8fdb\u884c\u52a8\u6001\u76ee\u6807\u5206\u5272\u6216\u5149\u6d41\u4f30\u8ba1\u3002\u6211\u4eec\u7684\u65b9\u6cd5\u5728\u4f20\u611f\u5668\u4eff\u771f\u4e2d\u5b9e\u73b0\u4e86\u6700\u5148\u8fdb\u7684\u6027\u80fd\uff0c\u5728\u91cd\u5efa\u9759\u6001\uff08+2.93 PSNR\uff09\u548c\u52a8\u6001\uff08+3.70 PSNR\uff09\u573a\u666f\u65f6\u660e\u663e\u4f18\u4e8e\u4ee5\u524d\u7684\u65b9\u6cd5\u3002\u6b64\u5916\uff0c\u4e3a\u4e86\u652f\u6301EmerNeRF\u7684\u8bed\u4e49\u6cdb\u5316\uff0c\u6211\u4eec\u5c062D\u89c6\u89c9\u57fa\u7840\u6a21\u578b\u7279\u5f81\u63d0\u5347\u52304D\u65f6\u7a7a\u4e2d\uff0c\u5e76\u89e3\u51b3\u4e86\u73b0\u4ee3\u53d8\u5f62\u91d1\u521a\u4e2d\u7684\u666e\u904d\u4f4d\u7f6e\u504f\u5dee\u95ee\u9898\uff0c\u663e\u8457\u63d0\u9ad8\u4e863D\u611f\u77e5\u6027\u80fd\uff08\u4f8b\u5982\uff0c\u804c\u4e1a\u9884\u6d4b\u7cbe\u5ea6\u5e73\u5747\u76f8\u5bf9\u63d0\u9ad8\u4e8637.50%\uff09\u3002\u6700\u540e\uff0c\u6211\u4eec\u6784\u5efa\u4e86\u4e00\u4e2a\u591a\u6837\u5316\u4e14\u5177\u6709\u6311\u6218\u6027\u7684120\u5e8f\u5217\u6570\u636e\u96c6\uff0c\u7528\u4e8e\u5728\u6781\u7aef\u548c\u9ad8\u5ea6\u52a8\u6001\u7684\u73af\u5883\u4e0b\u5bf9\u795e\u7ecf\u573a\u8fdb\u884c\u57fa\u51c6\u6d4b\u8bd5\u3002\u603b\u7ed3\u6765\u8bf4\uff0c\u672c\u6587\u7684\u4e3b\u8981\u8d21\u732e\u5982\u4e0b\uff1a<\/p>\n

              \n
            • EmerNeRF\u662f\u4e00\u79cd\u65b0\u9896\u76844D\u795e\u7ecf\u573a\u666f\u8868\u793a\u6846\u67b6\uff0c\u5728\u5177\u6709\u6311\u6218\u6027\u7684\u81ea\u52a8\u9a7e\u9a76\u573a\u666f\u4e2d\u8868\u73b0\u51fa\u8272\u3002EmerNeRF\u901a\u8fc7\u81ea\u76d1\u7763\u6267\u884c\u9759\u6001\u52a8\u6001\u5206\u89e3\u548c\u573a\u666f\u6d41\u4f30\u8ba1\uff1b<\/li>\n
            • \u4e00\u79cd\u7b80\u5316\u7684\u65b9\u6cd5\uff0c\u53ef\u4ee5\u89e3\u51b3ViT\u4e2d\u4f4d\u7f6e\u5d4c\u5165\u56fe\u6848\u7684\u4e0d\u826f\u5f71\u54cd\uff0c\u8be5\u65b9\u6cd5\u53ef\u7acb\u5373\u5e94\u7528\u4e8e\u5176\u4ed6\u4efb\u52a1;<\/li>\n
            • \u6211\u4eec\u5f15\u5165NOTR\u6570\u636e\u96c6\u6765\u8bc4\u4f30\u5404\u79cd\u6761\u4ef6\u4e0b\u7684\u795e\u7ecf\u573a\uff0c\u5e76\u4fc3\u8fdb\u8be5\u9886\u57df\u7684\u672a\u6765\u53d1\u5c55;<\/li>\n
            • EmerNeRF\u5728\u573a\u666f\u91cd\u5efa\u3001\u65b0\u89c6\u89d2\u5408\u6210\u548c\u573a\u666f\u6d41\u4f30\u8ba1\u65b9\u9762\u5b9e\u73b0\u4e86\u6700\u5148\u8fdb\u7684\u6027\u80fd\u3002<\/li>\n<\/ul>\n

              NeuRAD: Neural Rendering for Autonomous Driving<\/strong><\/p>\n

                \n
              • \u8bba\u6587\u94fe\u63a5\uff1ahttps:\/\/arxiv.org\/abs\/2311.15260v3<\/li>\n
              • \u4ee3\u7801\u94fe\u63a5\uff1ahttps:\/\/github.com\/georghess\/neurad-studio<\/li>\n<\/ul>\n

                \"\u56fe\u7247\"<\/p>\n

                Zenseact\u7684\u5de5\u4f5c\uff1a\u795e\u7ecf\u8f90\u5c04\u573a\uff08NeRF\uff09\u5728\u81ea\u52a8\u9a7e\u9a76\uff08AD\uff09\u9886\u57df\u8d8a\u6765\u8d8a\u53d7\u6b22\u8fce\u3002\u6700\u8fd1\u7684\u65b9\u6cd5\u8868\u660e\uff0cNeRF\u5177\u6709\u95ed\u73af\u4eff\u771f\u7684\u6f5c\u529b\uff0c\u80fd\u591f\u6d4b\u8bd5AD\u7cfb\u7edf\uff0c\u5e76\u4f5c\u4e3a\u4e00\u79cd\u5148\u8fdb\u7684\u8bad\u7ec3\u6570\u636e\u589e\u5f3a\u6280\u672f\u3002\u7136\u800c\uff0c\u73b0\u6709\u7684\u65b9\u6cd5\u901a\u5e38\u9700\u8981\u8f83\u957f\u7684\u8bad\u7ec3\u65f6\u95f4\u3001\u5bc6\u96c6\u7684\u8bed\u4e49\u76d1\u7763\u6216\u7f3a\u4e4f\u53ef\u63a8\u5e7f\u6027\u3002\u8fd9\u53cd\u8fc7\u6765\u53c8\u963b\u6b62\u4e86NeRFs\u5927\u89c4\u6a21\u5e94\u7528\u4e8eAD\u3002\u672c\u6587\u63d0\u51fa\u4e86NeuRAD\uff0c\u8fd9\u662f\u4e00\u79cd\u9488\u5bf9\u52a8\u6001AD\u6570\u636e\u91cf\u8eab\u5b9a\u5236\u7684\u9c81\u68d2\u65b0\u578b\u89c6\u56fe\u5408\u6210\u65b9\u6cd5\u3002\u6211\u4eec\u7684\u65b9\u6cd5\u5177\u6709\u7b80\u5355\u7684\u7f51\u7edc\u8bbe\u8ba1\uff0c\u5bf9\u76f8\u673a\u548c\u6fc0\u5149\u96f7\u8fbe\u8fdb\u884c\u4e86\u5e7f\u6cdb\u7684\u4f20\u611f\u5668\u5efa\u6a21\uff0c\u5305\u62ec\u6eda\u52a8\u5feb\u95e8\u3001\u5149\u675f\u53d1\u6563\u548c\u5149\u7ebf\u4e0b\u964d\uff0c\u9002\u7528\u4e8e\u5f00\u7bb1\u5373\u7528\u7684\u591a\u4e2a\u6570\u636e\u96c6\u3002\u6211\u4eec\u5728\u4e94\u4e2a\u6d41\u884c\u7684AD\u6570\u636e\u96c6\u4e0a\u9a8c\u8bc1\u4e86\u5b83\u7684\u6027\u80fd\uff0c\u5168\u9762\u5b9e\u73b0\u4e86\u6700\u5148\u8fdb\u7684\u6027\u80fd\u3002<\/p>\n

                DrivingGaussian: Composite Gaussian Splatting for Surrounding Dynamic Autonomous Driving Scenes<\/strong><\/p>\n

                  \n
                • \u8bba\u6587\u94fe\u63a5\uff1ahttps:\/\/arxiv.org\/abs\/2312.07920v3<\/li>\n
                • \u9879\u76ee\u4e3b\u9875\uff1ahttps:\/\/pkuvdig.github.io\/DrivingGaussian\/<\/li>\n<\/ul>\n

                  \"\u56fe\u7247\"<\/p>\n

                  \u5317\u5927&\u8c37\u6b4c\u7684\u5de5\u4f5c\uff1a\u672c\u6587\u63d0\u51fa\u4e86DrivingGaussian\u6a21\u578b\uff0c\u8fd9\u662f\u4e00\u4e2a\u7528\u4e8e\u73af\u89c6\u52a8\u6001\u81ea\u52a8\u9a7e\u9a76\u573a\u666f\u7684\u9ad8\u6548\u548c\u6709\u6548\u7684\u6846\u67b6\u3002\u5bf9\u4e8e\u5177\u6709\u8fd0\u52a8\u76ee\u6807\u7684\u590d\u6742\u573a\u666f\uff0cDrivingGaussian\u9996\u5148\u4f7f\u7528\u589e\u91cf\u9759\u60013D\u9ad8\u65af\u5bf9\u6574\u4e2a\u573a\u666f\u7684\u9759\u6001\u80cc\u666f\u8fdb\u884c\u987a\u5e8f\u548c\u6e10\u8fdb\u7684\u5efa\u6a21\u3002\u7136\u540e\u5229\u7528\u590d\u5408\u52a8\u6001\u9ad8\u65af\u56fe\u6765\u5904\u7406\u591a\u4e2a\u8fd0\u52a8\u76ee\u6807\uff0c\u5206\u522b\u91cd\u5efa\u6bcf\u4e2a\u76ee\u6807\u5e76\u6062\u590d\u5b83\u4eec\u5728\u573a\u666f\u4e2d\u7684\u51c6\u786e\u4f4d\u7f6e\u548c\u906e\u6321\u5173\u7cfb\u3002\u6211\u4eec\u8fdb\u4e00\u6b65\u4f7f\u7528\u6fc0\u5149\u96f7\u8fbe\u5148\u9a8c\u8fdb\u884c Gaussian Splatting\uff0c\u4ee5\u91cd\u5efa\u5177\u6709\u66f4\u591a\u7ec6\u8282\u7684\u573a\u666f\u5e76\u4fdd\u6301\u5168\u666f\u4e00\u81f4\u6027\u3002DrivingGaussian\u5728\u52a8\u6001\u9a71\u52a8\u573a\u666f\u91cd\u5efa\u65b9\u9762\u4f18\u4e8e\u73b0\u6709\u65b9\u6cd5\uff0c\u80fd\u591f\u5b9e\u73b0\u9ad8\u4fdd\u771f\u5ea6\u548c\u591a\u76f8\u673a\u4e00\u81f4\u6027\u7684\u903c\u771f\u73af\u7ed5\u89c6\u56fe\u5408\u6210\u3002\u603b\u7ed3\u6765\u8bf4\uff0c\u672c\u6587\u7684\u4e3b\u8981\u8d21\u732e\u5982\u4e0b\uff1a<\/p>\n

                    \n
                  • \u636e\u6211\u4eec\u6240\u77e5\uff0cDrivingGaussian\u662f\u57fa\u4e8e\u590d\u5408Gaussian Splatting\u7684\u5927\u89c4\u6a21\u52a8\u6001\u9a7e\u9a76\u573a\u666f\u7684\u7b2c\u4e00\u4e2a\u8868\u793a\u548c\u5efa\u6a21\u6846\u67b6\uff1b<\/li>\n
                  • \u5f15\u5165\u4e86\u4e24\u4e2a\u65b0\u6a21\u5757\uff0c\u5305\u62ec\u589e\u91cf\u9759\u60013D\u9ad8\u65af\u56fe\u548c\u590d\u5408\u52a8\u6001\u9ad8\u65af\u56fe\u3002\u524d\u8005\u9010\u6b65\u91cd\u5efa\u9759\u6001\u80cc\u666f\uff0c\u800c\u540e\u8005\u7528\u9ad8\u65af\u56fe\u5bf9\u591a\u4e2a\u52a8\u6001\u76ee\u6807\u8fdb\u884c\u5efa\u6a21\u3002\u5728\u6fc0\u5149\u96f7\u8fbe\u5148\u9a8c\u7684\u8f85\u52a9\u4e0b\uff0c\u6240\u63d0\u51fa\u7684\u65b9\u6cd5\u6709\u52a9\u4e8e\u5728\u5927\u89c4\u6a21\u9a7e\u9a76\u573a\u666f\u4e2d\u6062\u590d\u5b8c\u6574\u7684\u51e0\u4f55\u5f62\u72b6\uff1b<\/li>\n
                  • \u7efc\u5408\u5b9e\u9a8c\u8868\u660e\uff0cDriving Gaussian\u5728\u6311\u6218\u81ea\u52a8\u9a7e\u9a76\u57fa\u51c6\u6d4b\u8bd5\u65b9\u9762\u4f18\u4e8e\u4ee5\u524d\u7684\u65b9\u6cd5\uff0c\u5e76\u80fd\u591f\u4e3a\u5404\u79cd\u4e0b\u6e38\u4efb\u52a1\u8fdb\u884c\u89d2\u60c5\u51b5\u4eff\u771f\uff1b<\/li>\n<\/ul>\n

                    Street Gaussians: Modeling Dynamic Urban Scenes with Gaussian Splatting\uff08ECCV 2024\uff09<\/strong><\/p>\n

                      \n
                    • \u8bba\u6587\u94fe\u63a5\uff1ahttps:\/\/arxiv.org\/abs\/2401.01339v2<\/li>\n
                    • \u4ee3\u7801\u94fe\u63a5\uff1ahttps:\/\/github.com\/zju3dv\/street_gaussians<\/li>\n<\/ul>\n

                      \"\u56fe\u7247\"<\/p>\n

                      \u6d59\u5927&\u7406\u60f3\u5728ECCV 2024\u4e0a\u7684\u5de5\u4f5c\uff1a\u672c\u6587\u65e8\u5728\u89e3\u51b3\u81ea\u52a8\u9a7e\u9a76\u573a\u666f\u4e2d\u52a8\u6001\u57ce\u5e02\u8857\u9053\u7684\u5efa\u6a21\u95ee\u9898\u3002\u6700\u8fd1\u7684\u65b9\u6cd5\u901a\u8fc7\u5c06\u8ddf\u8e2a\u7684\u8f66\u8f86\u59ff\u6001\u7ed3\u5408\u5230\u8f66\u8f86\u52a8\u753b\u4e2d\u6765\u6269\u5c55NeRF\uff0c\u5b9e\u73b0\u4e86\u52a8\u6001\u57ce\u5e02\u8857\u9053\u573a\u666f\u7684\u7167\u7247\u7ea7\u903c\u771f\u89c6\u56fe\u5408\u6210\u3002\u7136\u800c\uff0c\u5b83\u4eec\u7684\u8bad\u7ec3\u901f\u5ea6\u548c\u6e32\u67d3\u901f\u5ea6\u90fd\u5f88\u6162\u3002\u4e3a\u6b64\u672c\u6587\u5f15\u5165\u4e86Street Gaussians\uff0c\u8fd9\u662f\u4e00\u79cd\u65b0\u7684\u663e\u5f0f\u573a\u666f\u8868\u793a\uff0c\u53ef\u4ee5\u89e3\u51b3\u8fd9\u4e9b\u9650\u5236\u3002\u5177\u4f53\u6765\u8bf4\uff0c\u52a8\u6001\u57ce\u5e02\u573a\u666f\u88ab\u8868\u793a\u4e3a\u4e00\u7ec4\u914d\u5907\u8bed\u4e49\u903b\u8f91\u548c3D\u9ad8\u65af\u7684\u70b9\u4e91\uff0c\u6bcf\u4e2a\u70b9\u4e91\u90fd\u4e0e\u524d\u666f\u8f66\u8f86\u6216\u80cc\u666f\u76f8\u5173\u8054\u3002\u4e3a\u4e86\u4eff\u771f\u524d\u666f\u76ee\u6807\u8f66\u8f86\u7684\u52a8\u529b\u5b66\uff0c\u6bcf\u4e2a\u76ee\u6807\u70b9\u4e91\u90fd\u4f7f\u7528\u53ef\u4f18\u5316\u7684\u8ddf\u8e2a\u59ff\u6001\u8fdb\u884c\u4f18\u5316\uff0c\u5e76\u4f7f\u75284D\u7403\u8c10\u6a21\u578b\u8fdb\u884c\u52a8\u6001\u5916\u89c2\u4f18\u5316\u3002\u663e\u5f0f\u8868\u793a\u5141\u8bb8\u8f7b\u677e\u7ec4\u5408\u76ee\u6807\u8f66\u8f86\u548c\u80cc\u666f\uff0c\u8fd9\u53cd\u8fc7\u6765\u53c8\u5141\u8bb8\u5728\u534a\u5c0f\u65f6\u7684\u8bad\u7ec3\u5185\u4ee5135 FPS\uff081066\u00d71600\u5206\u8fa8\u7387\uff09\u8fdb\u884c\u573a\u666f\u7f16\u8f91\u64cd\u4f5c\u548c\u6e32\u67d3\u3002\u8be5\u65b9\u6cd5\u5728\u591a\u4e2a\u5177\u6709\u6311\u6218\u6027\u7684\u57fa\u51c6\u4e0a\u8fdb\u884c\u4e86\u8bc4\u4f30\uff0c\u5305\u62ecKITTI\u548cWaymo Open\u6570\u636e\u96c6\u3002\u5b9e\u9a8c\u8868\u660e\u5728\u6240\u6709\u6570\u636e\u96c6\u4e0a\uff0c\u6240\u63d0\u51fa\u7684\u65b9\u6cd5\u59cb\u7ec8\u4f18\u4e8e\u6700\u5148\u8fdb\u7684\u65b9\u6cd5\u3002<\/p>\n

                      GaussianPro: 3D Gaussian Splatting with Progressive Propagation<\/strong><\/p>\n

                        \n
                      • \u8bba\u6587\u94fe\u63a5\uff1ahttps:\/\/arxiv.org\/abs\/2402.14650v1<\/li>\n
                      • \u4ee3\u7801\u94fe\u63a5\uff1ahttps:\/\/github.com\/kcheng1021\/GaussianPro<\/li>\n<\/ul>\n

                         <\/p>\n

                        \u4e2d\u79d1\u5927&\u6e2f\u5927\u7684\u5de5\u4f5c\uff1a3DGS\u7684\u51fa\u73b0\u6700\u8fd1\u5728\u795e\u7ecf\u6e32\u67d3\u9886\u57df\u5e26\u6765\u4e86\u4e00\u573a\u9769\u547d\uff0c\u4fc3\u8fdb\u4e86\u5b9e\u65f6\u901f\u5ea6\u7684\u9ad8\u8d28\u91cf\u6e32\u67d3\u3002\u7136\u800c\uff0c3DGS\u5728\u5f88\u5927\u7a0b\u5ea6\u4e0a\u4f9d\u8d56\u4e8e\u8fd0\u52a8\u7ed3\u6784\uff08SfM\uff09\u6280\u672f\u4ea7\u751f\u7684\u521d\u59cb\u5316\u70b9\u4e91\u3002\u5f53\u5904\u7406\u4e0d\u53ef\u907f\u514d\u5730\u5305\u542b\u65e0\u7eb9\u7406\u66f2\u9762\u7684\u5927\u89c4\u6a21\u573a\u666f\u65f6\uff0cSfM\u6280\u672f\u603b\u662f\u65e0\u6cd5\u5728\u8fd9\u4e9b\u66f2\u9762\u4e0a\u4ea7\u751f\u8db3\u591f\u7684\u70b9\uff0c\u4e5f\u65e0\u6cd5\u4e3a3DGS\u63d0\u4f9b\u826f\u597d\u7684\u521d\u59cb\u5316\u3002\u56e0\u6b64\uff0c3DGS\u5b58\u5728\u4f18\u5316\u56f0\u96be\u548c\u6e32\u67d3\u8d28\u91cf\u4f4e\u7684\u95ee\u9898\u3002\u5728\u8fd9\u7bc7\u8bba\u6587\u4e2d\uff0c\u53d7\u7ecf\u5178\u591a\u89c6\u56fe\u7acb\u4f53\uff08MVS\uff09\u6280\u672f\u7684\u542f\u53d1\uff0c\u6211\u4eec\u63d0\u51fa\u4e86GaussianPro\uff0c\u8fd9\u662f\u4e00\u79cd\u5e94\u7528\u6e10\u8fdb\u4f20\u64ad\u7b56\u7565\u6765\u6307\u5bfc3D Gaussian\u81f4\u5bc6\u5316\u7684\u65b0\u65b9\u6cd5\u3002\u4e0e3DGS\u4e2d\u4f7f\u7528\u7684\u7b80\u5355\u5206\u5272\u548c\u514b\u9686\u7b56\u7565\u76f8\u6bd4\uff0c\u6211\u4eec\u7684\u65b9\u6cd5\u5229\u7528\u573a\u666f\u73b0\u6709\u91cd\u5efa\u51e0\u4f55\u7684\u5148\u9a8c\u548c\u8865\u4e01\u5339\u914d\u6280\u672f\u6765\u751f\u6210\u5177\u6709\u7cbe\u786e\u4f4d\u7f6e\u548c\u65b9\u5411\u7684\u65b0\u9ad8\u65af\u5206\u5e03\u3002\u5728\u5927\u89c4\u6a21\u548c\u5c0f\u89c4\u6a21\u573a\u666f\u4e0a\u7684\u5b9e\u9a8c\u9a8c\u8bc1\u4e86\u6211\u4eec\u65b9\u6cd5\u7684\u6709\u6548\u6027\uff0c\u6211\u4eec\u7684\u65b9\u6cd5\u5728Waymo\u6570\u636e\u96c6\u4e0a\u663e\u8457\u8d85\u8fc7\u4e863DGS\uff0c\u5728PSNR\u65b9\u9762\u63d0\u9ad8\u4e861.15dB\u3002<\/p>\n

                        LidaRF: Delving into Lidar for Neural Radiance Field on Street Scenes<\/strong><\/p>\n

                          \n
                        • \u8bba\u6587\u94fe\u63a5\uff1ahttps:\/\/arxiv.org\/abs\/2405.00900v2<\/li>\n<\/ul>\n

                          \"\u56fe\u7247\"<\/p>\n

                          \u52a0\u5dde\u5927\u5b66\u6b27\u6587\u5206\u6821\u7684\u5de5\u4f5c\uff1a\u771f\u5b9e\u4eff\u771f\u5728\u81ea\u52a8\u9a7e\u9a76\u7b49\u5e94\u7528\u4e2d\u8d77\u7740\u81f3\u5173\u91cd\u8981\u7684\u4f5c\u7528\uff0c\u795e\u7ecf\u8f90\u5c04\u573a\uff08NeRF\uff09\u7684\u8fdb\u6b65\u53ef\u4ee5\u901a\u8fc7\u81ea\u52a8\u521b\u5efa\u6570\u5b573D\u8d44\u4ea7\u6765\u5b9e\u73b0\u66f4\u597d\u7684\u53ef\u6269\u5c55\u6027\u3002\u7136\u800c\uff0c\u7531\u4e8e\u5171\u7ebf\u76f8\u673a\u7684\u5927\u8fd0\u52a8\u548c\u9ad8\u901f\u4e0b\u7684\u7a00\u758f\u6837\u672c\uff0c\u8857\u9053\u573a\u666f\u7684\u91cd\u5efa\u8d28\u91cf\u4f1a\u53d7\u5230\u5f71\u54cd\u3002\u53e6\u4e00\u65b9\u9762\uff0c\u5b9e\u9645\u4f7f\u7528\u901a\u5e38\u8981\u6c42\u4ece\u504f\u79bb\u8f93\u5165\u7684\u76f8\u673a\u89c6\u56fe\u8fdb\u884c\u6e32\u67d3\uff0c\u4ee5\u51c6\u786e\u6a21\u62df\u8f66\u9053\u53d8\u6362\u7b49\u884c\u4e3a\u3002\u5728\u8fd9\u7bc7\u8bba\u6587\u4e2d\uff0c\u6211\u4eec\u63d0\u51fa\u4e86\u51e0\u4e2a\u89c1\u89e3\uff0c\u53ef\u4ee5\u66f4\u597d\u5730\u5229\u7528\u6fc0\u5149\u96f7\u8fbe\u6570\u636e\u6765\u63d0\u9ad8\u8857\u9053\u573a\u666f\u7684NeRF\u8d28\u91cf\u3002\u9996\u5148\uff0c\u6211\u4eec\u7684\u6846\u67b6\u4ece\u6fc0\u5149\u96f7\u8fbe\u4e2d\u5b66\u4e60\u51e0\u4f55\u573a\u666f\u8868\u793a\uff0c\u5c06\u5176\u4e0e\u9690\u5f0f\u57fa\u4e8e\u7f51\u683c\u7684\u8868\u793a\u878d\u5408\u7528\u4e8e\u8f90\u5c04\u89e3\u7801\uff0c\u7136\u540e\u63d0\u4f9b\u663e\u5f0f\u70b9\u4e91\u63d0\u4f9b\u7684\u66f4\u5f3a\u51e0\u4f55\u4fe1\u606f\u3002\u5176\u6b21\u63d0\u51fa\u4e86\u4e00\u79cd\u9c81\u68d2\u7684\u906e\u6321\u611f\u77e5\u6df1\u5ea6\u76d1\u7763\u65b9\u6848\uff0c\u8be5\u65b9\u6848\u5141\u8bb8\u901a\u8fc7\u7d2f\u79ef\u6765\u5229\u7528\u5bc6\u96c6\u7684\u6fc0\u5149\u96f7\u8fbe\u70b9\u3002\u7b2c\u4e09\u672c\u6587\u4ece\u6fc0\u5149\u96f7\u8fbe\u70b9\u751f\u6210\u589e\u5f3a\u8bad\u7ec3\u89c6\u56fe\uff0c\u4ee5\u8fdb\u4e00\u6b65\u6539\u8fdb\u3002\u6211\u4eec\u7684\u89c1\u89e3\u8f6c\u5316\u4e3a\u5728\u771f\u5b9e\u9a7e\u9a76\u573a\u666f\u4e0b\u5927\u5927\u6539\u8fdb\u7684\u65b0\u89c6\u56fe\u5408\u6210\u3002<\/p>\n

                          Gaussian: Self-Supervised Street Gaussians for Autonomous Driving<\/strong><\/p>\n

                            \n
                          • \u8bba\u6587\u94fe\u63a5\uff1ahttps:\/\/arxiv.org\/abs\/2405.20323v1<\/li>\n<\/ul>\n

                            \"\u56fe\u7247\"<\/p>\n

                            UC Berkeley&\u5317\u5927&\u6e05\u534e\u7684\u5de5\u4f5c\uff1a\u8857\u9053\u573a\u666f\u7684\u771f\u5b9e\u611f3D\u91cd\u5efa\u662f\u5f00\u53d1\u81ea\u52a8\u9a7e\u9a76\u4eff\u771f\u7684\u5173\u952e\u6280\u672f\u3002\u5c3d\u7ba1\u795e\u7ecf\u8f90\u5c04\u573a\uff08NeRF\uff09\u5728\u9a7e\u9a76\u573a\u666f\u4e2d\u7684\u6548\u7387\u5f88\u9ad8\uff0c\u4f463DGS\u56e0\u5176\u66f4\u5feb\u7684\u901f\u5ea6\u548c\u66f4\u660e\u786e\u7684\u8868\u793a\u800c\u6210\u4e3a\u4e00\u4e2a\u6709\u524d\u666f\u7684\u65b9\u5411\u3002\u7136\u800c\uff0c\u5927\u591a\u6570\u73b0\u6709\u7684\u8857\u90533DGS\u65b9\u6cd5\u9700\u8981\u8ddf\u8e2a\u76843D\u8f66\u8f86\u8fb9\u754c\u6846\u6765\u5206\u89e3\u9759\u6001\u548c\u52a8\u6001\u5143\u7d20\u4ee5\u8fdb\u884c\u6709\u6548\u7684\u91cd\u5efa\uff0c\u8fd9\u9650\u5236\u4e86\u5b83\u4eec\u5728\u81ea\u7531\u573a\u666f\u4e2d\u7684\u5e94\u7528\u3002\u4e3a\u4e86\u5728\u6ca1\u6709\u6807\u6ce8\u7684\u60c5\u51b5\u4e0b\u5b9e\u73b0\u9ad8\u6548\u76843D\u573a\u666f\u91cd\u5efa\uff0c\u6211\u4eec\u63d0\u51fa\u4e86\u4e00\u79cd\u81ea\u76d1\u7763\u8857\u9053\u9ad8\u65af\uff08S3Gaussian\uff09\u65b9\u6cd5\uff0c\u7528\u4e8e\u4ece4D\u4e00\u81f4\u6027\u4e2d\u5206\u89e3\u52a8\u6001\u548c\u9759\u6001\u5143\u7d20\u3002\u6211\u4eec\u75283D\u9ad8\u65af\u5206\u5e03\u6765\u8868\u793a\u6bcf\u4e2a\u573a\u666f\uff0c\u4ee5\u4fdd\u6301\u5176\u660e\u786e\u6027\uff0c\u5e76\u8fdb\u4e00\u6b65\u7528\u65f6\u7a7a\u573a\u7f51\u7edc\u6765\u538b\u7f294D\u52a8\u529b\u5b66\u6a21\u578b\u3002\u6211\u4eec\u5728\u5177\u6709\u6311\u6218\u6027\u7684Waymo Open\u6570\u636e\u96c6\u4e0a\u8fdb\u884c\u4e86\u5e7f\u6cdb\u7684\u5b9e\u9a8c\uff0c\u4ee5\u8bc4\u4f30\u6211\u4eec\u65b9\u6cd5\u7684\u6709\u6548\u6027\u3002\u6211\u4eec\u7684S3Gaussian\u5c55\u793a\u4e86\u5206\u89e3\u9759\u6001\u548c\u52a8\u6001\u573a\u666f\u7684\u80fd\u529b\uff0c\u5e76\u5728\u4e0d\u4f7f\u75283D\u6807\u6ce8\u7684\u60c5\u51b5\u4e0b\u5b9e\u73b0\u4e86\u6700\u4f73\u6027\u80fd\u3002<\/p>\n

                            Dynamic 3D Gaussian Fields for Urban Areas<\/strong><\/p>\n

                              \n
                            • \u8bba\u6587\u94fe\u63a5\uff1ahttps:\/\/arxiv.org\/abs\/2406.03175v1<\/li>\n
                            • \u4ee3\u7801\u94fe\u63a5\uff1ahttps:\/\/github.com\/tobiasfshr\/map4d\uff08\u5f85\u5f00\u6e90\uff09<\/li>\n<\/ul>\n

                              \"\u56fe\u7247\"<\/p>\n

                              ETH\u548cMeta\u7684\u5de5\u4f5c\uff1a\u672c\u6587\u63d0\u51fa\u4e86\u4e00\u79cd\u9ad8\u6548\u7684\u795e\u7ecf3D\u573a\u666f\u8868\u793a\u65b9\u6cd5\uff0c\u7528\u4e8e\u5927\u89c4\u6a21\u52a8\u6001\u57ce\u5e02\u5730\u533a\u7684\u65b0\u89c6\u56fe\u5408\u6210\uff08NVS\uff09\u3002\u7531\u4e8e\u5176\u6709\u9650\u7684\u89c6\u89c9\u8d28\u91cf\u548c\u975e\u4ea4\u4e92\u5f0f\u6e32\u67d3\u901f\u5ea6\uff0c\u73b0\u6709\u5de5\u4f5c\u54c1\u4e0d\u592a\u9002\u5408\u6df7\u5408\u73b0\u5b9e\u6216\u95ed\u73af\u4eff\u771f\u7b49\u5e94\u7528\u3002\u6700\u8fd1\uff0c\u57fa\u4e8e\u5149\u6805\u5316\u7684\u65b9\u6cd5\u4ee5\u4ee4\u4eba\u5370\u8c61\u6df1\u523b\u7684\u901f\u5ea6\u5b9e\u73b0\u4e86\u9ad8\u8d28\u91cf\u7684NVS\u3002\u7136\u800c\uff0c\u8fd9\u4e9b\u65b9\u6cd5\u4ec5\u9650\u4e8e\u5c0f\u89c4\u6a21\u3001\u5747\u5300\u7684\u6570\u636e\uff0c\u5373\u5b83\u4eec\u65e0\u6cd5\u5904\u7406\u7531\u4e8e\u5929\u6c14\u3001\u5b63\u8282\u548c\u5149\u7167\u5f15\u8d77\u7684\u4e25\u91cd\u5916\u89c2\u548c\u51e0\u4f55\u53d8\u5316\uff0c\u4e5f\u65e0\u6cd5\u6269\u5c55\u5230\u5177\u6709\u6570\u5343\u5f20\u56fe\u50cf\u7684\u66f4\u5927\u3001\u52a8\u6001\u7684\u533a\u57df\u3002\u6211\u4eec\u63d0\u51fa\u4e864DGF\uff0c\u8fd9\u662f\u4e00\u79cd\u795e\u7ecf\u573a\u666f\u8868\u793a\uff0c\u53ef\u6269\u5c55\u5230\u5927\u89c4\u6a21\u52a8\u6001\u57ce\u5e02\u533a\u57df\uff0c\u5904\u7406\u5f02\u6784\u8f93\u5165\u6570\u636e\uff0c\u5e76\u5927\u5927\u63d0\u9ad8\u4e86\u6e32\u67d3\u901f\u5ea6\u3002\u6211\u4eec\u4f7f\u75283D\u9ad8\u65af\u4f5c\u4e3a\u9ad8\u6548\u7684\u51e0\u4f55\u652f\u67b6\uff0c\u540c\u65f6\u4f9d\u8d56\u795e\u7ecf\u573a\u4f5c\u4e3a\u7d27\u51d1\u7075\u6d3b\u7684\u5916\u89c2\u6a21\u578b\u3002\u6211\u4eec\u901a\u8fc7\u5168\u5c40\u5c3a\u5ea6\u7684\u573a\u666f\u56fe\u96c6\u6210\u573a\u666f\u52a8\u529b\u5b66\uff0c\u540c\u65f6\u901a\u8fc7\u53d8\u5f62\u5728\u5c40\u90e8\u5c42\u9762\u5efa\u6a21\u5173\u8282\u8fd0\u52a8\u3002\u8fd9\u79cd\u5206\u89e3\u65b9\u6cd5\u5b9e\u73b0\u4e86\u9002\u7528\u4e8e\u73b0\u5b9e\u4e16\u754c\u5e94\u7528\u7684\u7075\u6d3b\u573a\u666f\u5408\u6210\u3002\u5728\u5b9e\u9a8c\u4e2d\uff0c\u6211\u4eec\u7ed5\u8fc7\u4e86\u6700\u5148\u8fdb\u7684\u6280\u672f\uff0cPSNR\u8d85\u8fc73dB\uff0c\u6e32\u67d3\u901f\u5ea6\u8d85\u8fc7200\u500d\u3002<\/p>\n

                              StreetSurf: Extending Multi-view Implicit Surface Reconstruction to Street Views<\/strong><\/p>\n

                                \n
                              • \u8bba\u6587\u94fe\u63a5\uff1ahttps:\/\/arxiv.org\/abs\/2306.04988v1<\/li>\n
                              • \u4ee3\u7801\u94fe\u63a5\uff1ahttps:\/\/github.com\/pjlab-ADG\/neuralsim<\/li>\n<\/ul>\n

                                \"\u56fe\u7247\"<\/p>\n

                                \u4e0a\u6d77AI Lab\u548c\u5546\u6c64\u7684\u5de5\u4f5c\uff1a\u672c\u6587\u63d0\u51fa\u4e86\u4e00\u79cd\u65b0\u7684\u591a\u89c6\u56fe\u9690\u5f0f\u8868\u9762\u91cd\u5efa\u6280\u672f\uff0c\u79f0\u4e3aStreetSurf\uff0c\u8be5\u6280\u672f\u5f88\u5bb9\u6613\u5e94\u7528\u4e8e\u5e7f\u6cdb\u4f7f\u7528\u7684\u81ea\u52a8\u9a7e\u9a76\u6570\u636e\u96c6\u4e2d\u7684\u8857\u666f\u56fe\u50cf\uff0c\u5982Waymo\u611f\u77e5\u5e8f\u5217\uff0c\u800c\u4e0d\u4e00\u5b9a\u9700\u8981LiDAR\u6570\u636e\u3002\u968f\u7740\u795e\u7ecf\u6e32\u67d3\u7814\u7a76\u7684\u8fc5\u901f\u53d1\u5c55\uff0c\u5c06\u5176\u6574\u5408\u5230\u8857\u666f\u4e2d\u5f00\u59cb\u5f15\u8d77\u4eba\u4eec\u7684\u5174\u8da3\u3002\u73b0\u6709\u7684\u8857\u666f\u65b9\u6cd5\u8981\u4e48\u4e3b\u8981\u5173\u6ce8\u65b0\u89c6\u56fe\u5408\u6210\uff0c\u5f88\u5c11\u63a2\u7d22\u573a\u666f\u51e0\u4f55\uff0c\u8981\u4e48\u5728\u7814\u7a76\u91cd\u5efa\u65f6\u4e25\u91cd\u4f9d\u8d56\u5bc6\u96c6\u7684LiDAR\u6570\u636e\u3002\u4ed6\u4eec\u90fd\u6ca1\u6709\u7814\u7a76\u591a\u89c6\u56fe\u9690\u5f0f\u8868\u9762\u91cd\u5efa\uff0c\u7279\u522b\u662f\u5728\u6ca1\u6709\u6fc0\u5149\u96f7\u8fbe\u6570\u636e\u7684\u60c5\u51b5\u4e0b\u3002\u6211\u4eec\u7684\u65b9\u6cd5\u6269\u5c55\u4e86\u73b0\u6709\u7684\u4ee5\u76ee\u6807\u4e3a\u4e2d\u5fc3\u7684\u795e\u7ecf\u8868\u9762\u91cd\u5efa\u6280\u672f\uff0c\u4ee5\u89e3\u51b3\u7531\u975e\u4ee5\u76ee\u6807\u4e3a\u6838\u5fc3\u3001\u957f\u800c\u7a84\u7684\u76f8\u673a\u8f68\u8ff9\u6355\u83b7\u7684\u65e0\u7ea6\u675f\u8857\u666f\u6240\u5e26\u6765\u7684\u72ec\u7279\u6311\u6218\u3002\u6211\u4eec\u5c06\u65e0\u7ea6\u675f\u7a7a\u95f4\u5212\u5206\u4e3a\u8fd1\u8ddd\u79bb\u3001\u8fdc\u666f\u548c\u5929\u7a7a\u4e09\u4e2a\u90e8\u5206\uff0c\u5177\u6709\u5bf9\u9f50\u7684\u957f\u65b9\u4f53\u8fb9\u754c\uff0c\u5e76\u91c7\u7528\u957f\u65b9\u4f53\/\u8d85\u957f\u65b9\u4f53\u54c8\u5e0c\u7f51\u683c\u4ee5\u53ca\u8def\u9762\u521d\u59cb\u5316\u65b9\u6848\uff0c\u4ee5\u5b9e\u73b0\u66f4\u7cbe\u7ec6\u548c\u66f4\u590d\u6742\u7684\u8868\u793a\u3002\u4e3a\u4e86\u8fdb\u4e00\u6b65\u89e3\u51b3\u65e0\u7eb9\u7406\u533a\u57df\u548c\u89c6\u89d2\u4e0d\u8db3\u5f15\u8d77\u7684\u51e0\u4f55\u8bef\u5dee\uff0c\u6211\u4eec\u91c7\u7528\u4e86\u4f7f\u7528\u901a\u7528\u5355\u76ee\u6a21\u578b\u4f30\u8ba1\u7684\u51e0\u4f55\u5148\u9a8c\u3002\u518d\u52a0\u4e0a\u6211\u4eec\u5b9e\u65bd\u4e86\u9ad8\u6548\u7ec6\u7c92\u5ea6\u7684\u591a\u7ea7\u5149\u7ebf\u884c\u8fdb\u7b56\u7565\uff0c\u6211\u4eec\u4f7f\u7528\u5355\u4e2aRTX3090 GPU\u5bf9\u6bcf\u4e2a\u8857\u9053\u89c6\u56fe\u5e8f\u5217\u8fdb\u884c\u8bad\u7ec3\uff0c\u4ec5\u9700\u4e00\u5230\u4e24\u4e2a\u5c0f\u65f6\u7684\u65f6\u95f4\uff0c\u5373\u53ef\u5728\u51e0\u4f55\u548c\u5916\u89c2\u65b9\u9762\u5b9e\u73b0\u6700\u5148\u8fdb\u7684\u91cd\u5efa\u8d28\u91cf\u3002\u6b64\u5916\uff0c\u6211\u4eec\u8bc1\u660e\u4e86\u91cd\u5efa\u7684\u9690\u5f0f\u66f2\u9762\u5728\u5404\u79cd\u4e0b\u6e38\u4efb\u52a1\u4e2d\u5177\u6709\u4e30\u5bcc\u7684\u6f5c\u529b\uff0c\u5305\u62ec\u5149\u7ebf\u8ffd\u8e2a\u548c\u6fc0\u5149\u96f7\u8fbe\u6a21\u62df\u3002<\/p>\n

                                AutoSplat: Constrained Gaussian Splatting for Autonomous Driving Scene Reconstruction<\/strong><\/p>\n

                                  \n
                                • \u8bba\u6587\u94fe\u63a5\uff1ahttps:\/\/arxiv.org\/abs\/2407.02598v2<\/li>\n<\/ul>\n

                                  \"\u56fe\u7247\"<\/p>\n

                                  \u591a\u4f26\u591a\u5927\u5b66\u548c\u534e\u4e3a\u8bfa\u4e9a\u7684\u5de5\u4f5c\uff1a\u903c\u771f\u7684\u573a\u666f\u91cd\u5efa\u548c\u89c6\u56fe\u5408\u6210\u5bf9\u4e8e\u901a\u8fc7\u4eff\u771f\u5b89\u5168\u5173\u952e\u573a\u666f\u6765\u63a8\u8fdb\u81ea\u52a8\u9a7e\u9a76\u7cfb\u7edf\u81f3\u5173\u91cd\u8981\u30023DGS\u5728\u5b9e\u65f6\u6e32\u67d3\u548c\u9759\u6001\u573a\u666f\u91cd\u5efa\u65b9\u9762\u8868\u73b0\u51fa\u8272\uff0c\u4f46\u7531\u4e8e\u590d\u6742\u7684\u80cc\u666f\u3001\u52a8\u6001\u5bf9\u8c61\u548c\u7a00\u758f\u89c6\u56fe\uff0c\u5728\u5efa\u6a21\u9a7e\u9a76\u573a\u666f\u65b9\u9762\u9047\u5230\u4e86\u56f0\u96be\u3002\u6211\u4eec\u63d0\u51fa\u4e86AutoPlat\uff0c\u8fd9\u662f\u4e00\u4e2a\u91c7\u7528Gaussian Splatting\u5b9e\u73b0\u81ea\u52a8\u9a7e\u9a76\u573a\u666f\u9ad8\u5ea6\u903c\u771f\u91cd\u5efa\u7684\u6846\u67b6\u3002\u901a\u8fc7\u5bf9\u8868\u793a\u9053\u8def\u548c\u5929\u7a7a\u533a\u57df\u7684\u9ad8\u65af\u5206\u5e03\u56fe\u65bd\u52a0\u51e0\u4f55\u7ea6\u675f\uff0c\u6211\u4eec\u7684\u65b9\u6cd5\u80fd\u591f\u5bf9\u5305\u62ec\u8f66\u9053\u53d8\u6362\u5728\u5185\u7684\u5177\u6709\u6311\u6218\u6027\u7684\u573a\u666f\u8fdb\u884c\u591a\u89c6\u56fe\u4e00\u81f4\u7684\u6a21\u62df\u3002\u5229\u75283D\u6a21\u677f\uff0c\u6211\u4eec\u5f15\u5165\u4e86\u53cd\u5c04\u9ad8\u65af\u4e00\u81f4\u6027\u7ea6\u675f\u6765\u76d1\u7763\u524d\u666f\u5bf9\u8c61\u7684\u53ef\u89c1\u9762\u548c\u4e0d\u53ef\u89c1\u9762\u3002\u6b64\u5916\uff0c\u4e3a\u4e86\u6a21\u62df\u524d\u666f\u5bf9\u8c61\u7684\u52a8\u6001\u5916\u89c2\uff0c\u6211\u4eec\u4f30\u8ba1\u4e86\u6bcf\u4e2a\u524d\u666f\u9ad8\u65af\u7684\u6b8b\u5dee\u7403\u9762\u8c10\u6ce2\u3002\u5728Pandaset\u548cKITTI\u4e0a\u8fdb\u884c\u7684\u5927\u91cf\u5b9e\u9a8c\u8868\u660e\uff0cAutoPlat\u5728\u5404\u79cd\u9a7e\u9a76\u573a\u666f\u4e2d\u7684\u573a\u666f\u91cd\u5efa\u548c\u65b0\u9896\u89c6\u56fe\u5408\u6210\u65b9\u9762\u4f18\u4e8e\u6700\u5148\u8fdb\u7684\u65b9\u6cd5\u3002<\/p>\n

                                  DHGS: Decoupled Hybrid Gaussian Splatting for Driving Scene<\/strong><\/p>\n

                                    \n
                                  • \u8bba\u6587\u94fe\u63a5\uff1ahttps:\/\/arxiv.org\/abs\/2407.16600v3<\/li>\n<\/ul>\n

                                    \"\u56fe\u7247\"<\/p>\n

                                    \u957f\u5b89\u6c7d\u8f66\u7684\u5de5\u4f5c\uff1a\u73b0\u6709\u7684GS\u65b9\u6cd5\u5728\u5b9e\u73b0\u9a7e\u9a76\u573a\u666f\u4e2d\u4ee4\u4eba\u6ee1\u610f\u7684\u65b0\u89c6\u56fe\u5408\u6210\u65b9\u9762\u5f80\u5f80\u4e0d\u8db3\uff0c\u4e3b\u8981\u662f\u7531\u4e8e\u7f3a\u4e4f\u5de7\u5999\u7684\u8bbe\u8ba1\u548c\u6240\u6d89\u53ca\u5143\u7d20\u7684\u51e0\u4f55\u7ea6\u675f\u3002\u672c\u6587\u4ecb\u7ecd\u4e86\u4e00\u79cd\u65b0\u7684\u795e\u7ecf\u6e32\u67d3\u65b9\u6cd5\uff0c\u79f0\u4e3a\u89e3\u8026\u6df7\u5408GS\uff08DHGS\uff09\uff0c\u65e8\u5728\u63d0\u9ad8\u9759\u6001\u9a7e\u9a76\u573a\u666f\u65b0\u578b\u89c6\u56fe\u5408\u6210\u7684\u6e32\u67d3\u8d28\u91cf\u3002\u8fd9\u9879\u5de5\u4f5c\u7684\u65b0\u9896\u4e4b\u5904\u5728\u4e8e\uff0c\u9488\u5bf9\u9053\u8def\u548c\u975e\u9053\u8def\u5c42\u7684\u89e3\u8026\u548c\u6df7\u5408\u50cf\u7d20\u7ea7\u6df7\u5408\u5668\uff0c\u6ca1\u6709\u9488\u5bf9\u6574\u4e2a\u573a\u666f\u7684\u4f20\u7edf\u7edf\u4e00\u5dee\u5206\u6e32\u67d3\u903b\u8f91\uff0c\u540c\u65f6\u901a\u8fc7\u63d0\u51fa\u7684\u6df1\u5ea6\u6709\u5e8f\u6df7\u5408\u6e32\u67d3\u7b56\u7565\u4ecd\u7136\u4fdd\u6301\u4e00\u81f4\u548c\u8fde\u7eed\u7684\u53e0\u52a0\u3002\u6b64\u5916\uff0c\u5bf9\u7531\u7b26\u53f7\u8ddd\u79bb\u573a\uff08SDF\uff09\u7ec4\u6210\u7684\u9690\u5f0f\u9053\u8def\u8868\u793a\u8fdb\u884c\u8bad\u7ec3\uff0c\u4ee5\u76d1\u63a7\u5177\u6709\u5fae\u5999\u51e0\u4f55\u5c5e\u6027\u7684\u8def\u9762\u3002\u4f34\u968f\u7740\u8f85\u52a9\u4f20\u8f93\u635f\u8017\u548c\u4e00\u81f4\u6027\u635f\u8017\u7684\u4f7f\u7528\uff0c\u6700\u7ec8\u4fdd\u7559\u4e86\u5177\u6709\u4e0d\u53ef\u5bdf\u89c9\u8fb9\u754c\u548c\u9ad8\u4fdd\u771f\u5ea6\u7684\u65b0\u56fe\u50cf\u3002\u5728Waymo\u6570\u636e\u96c6\u4e0a\u8fdb\u884c\u7684\u5927\u91cf\u5b9e\u9a8c\u8bc1\u660e\uff0cDHGS\u7684\u6027\u80fd\u4f18\u4e8e\u6700\u5148\u8fdb\u7684\u65b9\u6cd5\u3002<\/p>\n<\/div>\n<\/div>\n

                                    <\/div>\n

                                    <\/g><\/svg><\/i> \"Loading\"<\/p>\n

                                    <\/div>\n","protected":false},"excerpt":{"rendered":"

                                    \u4eca\u5929\u4e3a\u5927\u5bb6\u76d8\u70b93DGS&NeRF\u5728\u81ea\u52a8\u9a7e\u9a76\u91cd\u5efa\u65b9\u9762\u7684\u5de5\u4f5c\uff01 \u672c\u6587\u7ecf\u81ea\u52a8\u9a7e\u9a76\u4e4b\u5fc3\u516c\u4f17\u53f7\u6388\u6743\u8f6c\u8f7d\uff0c\u8f6c\u8f7d\u8bf7\u8054 […]<\/p>\n

                                    <\/div>\n

                                    <\/g><\/svg><\/i> \"Loading\"<\/p>\n

                                    <\/div>\n","protected":false},"author":56,"featured_media":21947,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[23,20,80],"tags":[1280],"class_list":["post-21945","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-23","category-20","category-80","tag-3d-"],"_links":{"self":[{"href":"https:\/\/aif.amtbbs.org\/index.php\/wp-json\/wp\/v2\/posts\/21945","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/aif.amtbbs.org\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/aif.amtbbs.org\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/aif.amtbbs.org\/index.php\/wp-json\/wp\/v2\/users\/56"}],"replies":[{"embeddable":true,"href":"https:\/\/aif.amtbbs.org\/index.php\/wp-json\/wp\/v2\/comments?post=21945"}],"version-history":[{"count":1,"href":"https:\/\/aif.amtbbs.org\/index.php\/wp-json\/wp\/v2\/posts\/21945\/revisions"}],"predecessor-version":[{"id":21948,"href":"https:\/\/aif.amtbbs.org\/index.php\/wp-json\/wp\/v2\/posts\/21945\/revisions\/21948"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/aif.amtbbs.org\/index.php\/wp-json\/wp\/v2\/media\/21947"}],"wp:attachment":[{"href":"https:\/\/aif.amtbbs.org\/index.php\/wp-json\/wp\/v2\/media?parent=21945"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/aif.amtbbs.org\/index.php\/wp-json\/wp\/v2\/categories?post=21945"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/aif.amtbbs.org\/index.php\/wp-json\/wp\/v2\/tags?post=21945"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}