{"version":2,"relationshipURLBase":"https://codingai-lec03.pages.dev/course/notebook-blocks/365db128e760487bbb6de81c4068f3a7.html","relationshipResolution":"Resolve relative relationship URLs against relationshipURLBase; complete records and absolute URLs are in blocks.","title":"来源与结构输出的连接","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/365db128e760487bbb6de81c4068f3a7.html","markdown":"https://codingai-lec03.pages.dev/course/notebook-blocks/365db128e760487bbb6de81c4068f3a7.md","pages":[{"id":"reading-opening","title":"来源与结构输出的连接","unit":"来源与结构输出的连接","unitId":"notebook-blocks/365db128e760487bbb6de81c4068f3a7","chapter":"course","number":1}],"aliases":[{"oldId":"b-383a6104db48e43d1eaa0295","targetId":"b-383a6104db48e43d1eaa0295","page":"reading-opening","operation":"move","beforeSHA256":"f6dfd09a50caab375f02e0abcc3ff7a454ea579ed8eef0c35b95c95b0619c304","afterSHA256":"f6dfd09a50caab375f02e0abcc3ff7a454ea579ed8eef0c35b95c95b0619c304","reason":"Same source body in authored semantic page"},{"oldId":"b-8627debb04e6e3136192db4b","targetId":"b-8627debb04e6e3136192db4b","page":"reading-opening","operation":"move","beforeSHA256":"34c169c4b521c9c3c678b54e09c87b13842cf758093dcf67948c4c3dc9b8297b","afterSHA256":"34c169c4b521c9c3c678b54e09c87b13842cf758093dcf67948c4c3dc9b8297b","reason":"Same source body in authored semantic page"},{"oldId":"b-a1171cd1b866740f190bb768","targetId":"b-a1171cd1b866740f190bb768","page":"reading-opening","operation":"move","beforeSHA256":"e91f24e4e4ca40e2099a0d81d30db6568f4bf2074b499eeee6e3959617bd97ac","afterSHA256":"e91f24e4e4ca40e2099a0d81d30db6568f4bf2074b499eeee6e3959617bd97ac","reason":"Same source body in authored semantic page"},{"oldId":"b-3206a00e2aa2b48e530de56c","targetId":"b-3206a00e2aa2b48e530de56c","page":"reading-opening","operation":"move","beforeSHA256":"6f0b115c3c32032428a4c0d5ec4791e16d6b63aa3d643457bed758bf6af2b765","afterSHA256":"6f0b115c3c32032428a4c0d5ec4791e16d6b63aa3d643457bed758bf6af2b765","reason":"Same source body in authored semantic page"},{"oldId":"b-2c65775c27be172d61247094","targetId":"b-2c65775c27be172d61247094","page":"reading-opening","operation":"move","beforeSHA256":"96b5617dd1b75d91c269215ecedcbf9a577b0f5f72047a0da9474c2f4901f298","afterSHA256":"96b5617dd1b75d91c269215ecedcbf9a577b0f5f72047a0da9474c2f4901f298","reason":"Same source body in authored semantic page"},{"oldId":"b-c3b688b179aca32485bd463f","targetId":"b-c3b688b179aca32485bd463f","page":"reading-opening","operation":"move","beforeSHA256":"704681ab580f1aba8abf702f224a6393f184b62c6bc8c2aebd06660b7039e0a8","afterSHA256":"704681ab580f1aba8abf702f224a6393f184b62c6bc8c2aebd06660b7039e0a8","reason":"Same source body in authored semantic page"},{"oldId":"b-9b43420f9b67d6eaf6b17abb","targetId":"b-9b43420f9b67d6eaf6b17abb","page":"reading-opening","operation":"move","beforeSHA256":"986aeb0d9133f40c4ba6c833a309426dce10edbd3c280d3d5f7969b14e818b23","afterSHA256":"986aeb0d9133f40c4ba6c833a309426dce10edbd3c280d3d5f7969b14e818b23","reason":"Same source body in authored semantic page"},{"oldId":"block-365db128e760487bbb6de81c4068f3a7--来源与结构输出的连接","targetId":"b-2c65775c27be172d61247094","operation":"retain-source-identity"}],"blocks":[{"id":"b-383a6104db48e43d1eaa0295","type":"p","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/365db128e760487bbb6de81c4068f3a7.html#b-383a6104db48e43d1eaa0295","content":"[LAB 11 · 从 RGB 到像素：读出、FCN 与边界](/course/notebooks/9e268706471944e3a59e6f568f9ce9a3.html)","sha256":"f6dfd09a50caab375f02e0abcc3ff7a454ea579ed8eef0c35b95c95b0619c304","chapter":{"id":"course","title":"来源与结构输出的连接"},"unit":{"id":"notebook-blocks/365db128e760487bbb6de81c4068f3a7","title":"来源与结构输出的连接"},"page":{"id":"reading-opening","title":"来源与结构输出的连接","number":1},"sectionTitle":"来源与结构输出的连接","currentQuestion":"来源与结构输出的连接","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":null,"nextBlock":"b-8627debb04e6e3136192db4b","prerequisites":[],"relatedBlocks":[]},{"id":"b-8627debb04e6e3136192db4b","type":"h1","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/365db128e760487bbb6de81c4068f3a7.html#b-8627debb04e6e3136192db4b","content":"# 来源与结构输出的连接","sha256":"34c169c4b521c9c3c678b54e09c87b13842cf758093dcf67948c4c3dc9b8297b","chapter":{"id":"course","title":"来源与结构输出的连接"},"unit":{"id":"notebook-blocks/365db128e760487bbb6de81c4068f3a7","title":"来源与结构输出的连接"},"page":{"id":"reading-opening","title":"来源与结构输出的连接","number":1},"sectionTitle":"来源与结构输出的连接","currentQuestion":"来源与结构输出的连接","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-383a6104db48e43d1eaa0295","nextBlock":"b-a1171cd1b866740f190bb768","prerequisites":[{"id":"b-383a6104db48e43d1eaa0295","url":"#b-383a6104db48e43d1eaa0295","role":"problem-context"}],"relatedBlocks":[]},{"id":"b-a1171cd1b866740f190bb768","type":"p","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/365db128e760487bbb6de81c4068f3a7.html#b-a1171cd1b866740f190bb768","content":"按问题继续","sha256":"e91f24e4e4ca40e2099a0d81d30db6568f4bf2074b499eeee6e3959617bd97ac","chapter":{"id":"course","title":"来源与结构输出的连接"},"unit":{"id":"notebook-blocks/365db128e760487bbb6de81c4068f3a7","title":"来源与结构输出的连接"},"page":{"id":"reading-opening","title":"来源与结构输出的连接","number":1},"sectionTitle":"来源与结构输出的连接","currentQuestion":"来源与结构输出的连接","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-8627debb04e6e3136192db4b","nextBlock":"b-3206a00e2aa2b48e530de56c","prerequisites":[{"id":"b-8627debb04e6e3136192db4b","url":"#b-8627debb04e6e3136192db4b","role":"section-title"},{"id":"b-383a6104db48e43d1eaa0295","url":"#b-383a6104db48e43d1eaa0295","role":"problem-context"}],"relatedBlocks":[]},{"id":"b-3206a00e2aa2b48e530de56c","type":"p","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/365db128e760487bbb6de81c4068f3a7.html#b-3206a00e2aa2b48e530de56c","content":"前景/背景/mIoU口径、RGB到mask训练、FCN与1×1读出及边界","sha256":"6f0b115c3c32032428a4c0d5ec4791e16d6b63aa3d643457bed758bf6af2b765","chapter":{"id":"course","title":"来源与结构输出的连接"},"unit":{"id":"notebook-blocks/365db128e760487bbb6de81c4068f3a7","title":"来源与结构输出的连接"},"page":{"id":"reading-opening","title":"来源与结构输出的连接","number":1},"sectionTitle":"来源与结构输出的连接","currentQuestion":"来源与结构输出的连接","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-a1171cd1b866740f190bb768","nextBlock":"b-2c65775c27be172d61247094","prerequisites":[{"id":"b-8627debb04e6e3136192db4b","url":"#b-8627debb04e6e3136192db4b","role":"section-title"},{"id":"b-383a6104db48e43d1eaa0295","url":"#b-383a6104db48e43d1eaa0295","role":"problem-context"}],"relatedBlocks":[]},{"id":"b-2c65775c27be172d61247094","type":"h2","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/365db128e760487bbb6de81c4068f3a7.html#b-2c65775c27be172d61247094","content":"## 来源与结构输出的连接","sha256":"96b5617dd1b75d91c269215ecedcbf9a577b0f5f72047a0da9474c2f4901f298","chapter":{"id":"course","title":"来源与结构输出的连接"},"unit":{"id":"notebook-blocks/365db128e760487bbb6de81c4068f3a7","title":"来源与结构输出的连接"},"page":{"id":"reading-opening","title":"来源与结构输出的连接","number":1},"sectionTitle":"来源与结构输出的连接","currentQuestion":"来源与结构输出的连接","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-3206a00e2aa2b48e530de56c","nextBlock":"b-c3b688b179aca32485bd463f","prerequisites":[{"id":"b-8627debb04e6e3136192db4b","url":"#b-8627debb04e6e3136192db4b","role":"section-title"},{"id":"b-383a6104db48e43d1eaa0295","url":"#b-383a6104db48e43d1eaa0295","role":"problem-context"}],"relatedBlocks":[]},{"id":"b-c3b688b179aca32485bd463f","type":"p","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/365db128e760487bbb6de81c4068f3a7.html#b-c3b688b179aca32485bd463f","content":"2026-09-22 实时核验 [D2L §14.11 Fully Convolutional Networks](https://d2l.ai/chapter_computer-vision/fcn.html)，采用全卷积逐像素读出、上采样和双线性坐标的解释；本实现使用可学习粗/细分支加双线性插值，没有预训练骨干，也没有宣称复现其 VOC 实验。ignore、三类混淆矩阵、两个颜色场景和条件对照是本课程的独立教学构造。","sha256":"704681ab580f1aba8abf702f224a6393f184b62c6bc8c2aebd06660b7039e0a8","chapter":{"id":"course","title":"来源与结构输出的连接"},"unit":{"id":"notebook-blocks/365db128e760487bbb6de81c4068f3a7","title":"来源与结构输出的连接"},"page":{"id":"reading-opening","title":"来源与结构输出的连接","number":1},"sectionTitle":"来源与结构输出的连接","currentQuestion":"来源与结构输出的连接","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-2c65775c27be172d61247094","nextBlock":"b-9b43420f9b67d6eaf6b17abb","prerequisites":[{"id":"b-2c65775c27be172d61247094","url":"#b-2c65775c27be172d61247094","role":"section-title"},{"id":"b-383a6104db48e43d1eaa0295","url":"#b-383a6104db48e43d1eaa0295","role":"problem-context"}],"relatedBlocks":[]},{"id":"b-9b43420f9b67d6eaf6b17abb","type":"p","scope":"document","url":"https://codingai-lec03.pages.dev/course/notebook-blocks/365db128e760487bbb6de81c4068f3a7.html#b-9b43420f9b67d6eaf6b17abb","content":"LAB10 中检测框正确只说明包围区域满足评价阈值。LAB11 中椭圆边缘、相邻对象间的背景缝、未标注像素仍是不同问题：输出粒度决定什么算错误，采样与任务头决定哪些空间信息能留下。返回网页的空间主线时，可以用本册三张同源输入/目标/错误图说明这一步转折，而不只说“语义分割更精细”。","sha256":"986aeb0d9133f40c4ba6c833a309426dce10edbd3c280d3d5f7969b14e818b23","chapter":{"id":"course","title":"来源与结构输出的连接"},"unit":{"id":"notebook-blocks/365db128e760487bbb6de81c4068f3a7","title":"来源与结构输出的连接"},"page":{"id":"reading-opening","title":"来源与结构输出的连接","number":1},"sectionTitle":"来源与结构输出的连接","currentQuestion":"来源与结构输出的连接","sourceSlides":[],"knowledgeBlock":null,"formulas":[],"previousBlock":"b-c3b688b179aca32485bd463f","nextBlock":null,"prerequisites":[{"id":"b-2c65775c27be172d61247094","url":"#b-2c65775c27be172d61247094","role":"section-title"},{"id":"b-383a6104db48e43d1eaa0295","url":"#b-383a6104db48e43d1eaa0295","role":"problem-context"}],"relatedBlocks":[]}]}