[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6KFNFA1B47H7XEMGNZK35VT":3,"skill-same:01M6KFNFA1B47H7XEMGNZK35VT":59},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":32,"linkOnly":33,"markdownContent":34,"markdownBody":35,"frontMatter":36,"refName":37,"commitSha":38,"skillDirectory":39,"forksCount":40,"repositoryHtmlUrl":41,"repositoryHomepageUrl":42,"ownerHtmlUrl":43,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":44,"install":45,"resources":58},"01M6KFNFA1B47H7XEMGNZK35VT","run2_template-matching-tuning",null,"Tune OpenCV template matching parameters (threshold and dedup distance) for accurate object counting in pixel-art games.","cxcscmu\u002FSkillLearnBench","cxcscmu","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F142838272?v=4","01M6KCQDFQWPDSN0MKZY96MSQP","01M6KCQDM7G70TKJ28NTSQC0RZ",93,"MIT","https:\u002F\u002Fgithub.com\u002Fcxcscmu\u002FSkillLearnBench\u002Fblob\u002Fmain\u002Fskills\u002Fb2-self-feedback-claude-opus-4-6\u002Fvideo-object-counting\u002Frun2_template-matching-tuning\u002FSKILL.md",[],[18,19,20,21,22,23,24,25,26,27,28,29,30,31],"tuning","template","matching","run2 template matching tuning","opencv","Art","pixel","counting","tune","games","pixel-art","object","for","opencv template matching","2026-11-10 10:12:30",false,"---\nname: run2_template-matching-tuning\ndescription: Tune OpenCV template matching parameters (threshold and dedup distance) for accurate object counting in pixel-art games.\n---\n\n# Template Matching Parameter Tuning for Pixel-Art Games\n\n## Tool\n\n```bash\npython3 \u002Froot\u002F.claude\u002Fskills\u002Fobject_counter\u002Fscripts\u002Fcount_objects.py \\\n    --tool count \\\n    --input_image \u003Cframe.png> \\\n    --object_image \u003Ctemplate.png> \\\n    --threshold \u003Cfloat> \\\n    --dedup_min_dist \u003Cfloat>\n```\n\n## Parameter Tuning Strategy\n\n### Threshold\n- **0.9** (default): Strict matching. Works when objects appear identical to template.\n- **0.75-0.85**: Better for pixel-art games where objects have animation frames (e.g., spinning coins show different faces).\n- **Below 0.7**: Too many false positives from similar-colored regions.\n\n### dedup_min_dist\n- **3** (default): Works for very small templates but can double-count larger objects.\n- **5-10**: Better for game sprites (~16-32px wide). Prevents overlapping detections.\n\n## Recommended Parameters for Super Mario\n\n| Object  | Threshold | dedup_min_dist | Rationale |\n|---------|-----------|----------------|-----------|\n| Coin    | 0.75      | 5              | Coins have multiple animation frames; lower threshold catches rotated coins |\n| Enemy   | 0.9       | 5              | Goombas are distinctive and consistent in appearance |\n| Turtle  | 0.9       | 5              | Koopa turtles are distinctive; 0.9 avoids false positives from terrain |\n\n## Tuning Process\n\n1. Start with threshold=0.9, dedup=3 (recommended defaults)\n2. Visually inspect each frame to count expected objects\n3. Sweep thresholds (0.9, 0.85, 0.8, 0.75, 0.7) while checking for:\n   - False negatives (objects missed → lower threshold)\n   - False positives (non-objects counted → raise threshold or dedup_min_dist)\n4. Increase dedup_min_dist if duplicate detections appear at lower thresholds\n5. Different object types may need different optimal thresholds\n\n## Key Insight\n\nThe HUD coin icon (top of screen) is very small and typically doesn't match the template at normal thresholds. Game coins in the play area are larger and match well.\n","# Template Matching Parameter Tuning for Pixel-Art Games\n\n## Tool\n\n```bash\npython3 \u002Froot\u002F.claude\u002Fskills\u002Fobject_counter\u002Fscripts\u002Fcount_objects.py \\\n    --tool count \\\n    --input_image \u003Cframe.png> \\\n    --object_image \u003Ctemplate.png> \\\n    --threshold \u003Cfloat> \\\n    --dedup_min_dist \u003Cfloat>\n```\n\n## Parameter Tuning Strategy\n\n### Threshold\n- **0.9** (default): Strict matching. Works when objects appear identical to template.\n- **0.75-0.85**: Better for pixel-art games where objects have animation frames (e.g., spinning coins show different faces).\n- **Below 0.7**: Too many false positives from similar-colored regions.\n\n### dedup_min_dist\n- **3** (default): Works for very small templates but can double-count larger objects.\n- **5-10**: Better for game sprites (~16-32px wide). Prevents overlapping detections.\n\n## Recommended Parameters for Super Mario\n\n| Object  | Threshold | dedup_min_dist | Rationale |\n|---------|-----------|----------------|-----------|\n| Coin    | 0.75      | 5              | Coins have multiple animation frames; lower threshold catches rotated coins |\n| Enemy   | 0.9       | 5              | Goombas are distinctive and consistent in appearance |\n| Turtle  | 0.9       | 5              | Koopa turtles are distinctive; 0.9 avoids false positives from terrain |\n\n## Tuning Process\n\n1. Start with threshold=0.9, dedup=3 (recommended defaults)\n2. Visually inspect each frame to count expected objects\n3. Sweep thresholds (0.9, 0.85, 0.8, 0.75, 0.7) while checking for:\n   - False negatives (objects missed → lower threshold)\n   - False positives (non-objects counted → raise threshold or dedup_min_dist)\n4. Increase dedup_min_dist if duplicate detections appear at lower thresholds\n5. Different object types may need different optimal thresholds\n\n## Key Insight\n\nThe HUD coin icon (top of screen) is very small and typically doesn't match the template at normal thresholds. Game coins in the play area are larger and match well.\n",{"name":5,"description":7},"main","726e1f79b96b0cb3d729d0e55a1c847e0b5353ae","skills\u002Fb2-self-feedback-claude-opus-4-6\u002Fvideo-object-counting\u002Frun2_template-matching-tuning",8,"https:\u002F\u002Fgithub.com\u002Fcxcscmu\u002FSkillLearnBench","https:\u002F\u002Fcxcscmu.github.io\u002FSkillLearnBench","https:\u002F\u002Fgithub.com\u002Fcxcscmu","2026-10-02 22:55:36",{"source":8,"skillName":5,"commands":46,"prompt":56,"zipAvailable":57},[47,50,53],{"tool":48,"command":49},"npx","npx skills add cxcscmu\u002FSkillLearnBench --skill run2_template-matching-tuning",{"tool":51,"command":52},"bunx","bunx skills add cxcscmu\u002FSkillLearnBench --skill run2_template-matching-tuning",{"tool":54,"command":55},"pnpm","pnpm dlx skills add cxcscmu\u002FSkillLearnBench --skill run2_template-matching-tuning","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fcxcscmu\u002FSkillLearnBench\u002Fblob\u002Fmain\u002Fskills\u002Fb2-self-feedback-claude-opus-4-6\u002Fvideo-object-counting\u002Frun2_template-matching-tuning\u002FSKILL.md",true,[],[60,77,101,119,138,150],{"skillNo":61,"name":62,"nameZh":6,"description":63,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":64,"categories":65,"keywords":71,"publishedAt":32},"01M6KM5KH175P5ABP5JMX2SBDQ","object_counter","Count occurrences of an object in the image using computer vision algorithm.","https:\u002F\u002Fgithub.com\u002Fcxcscmu\u002FSkillLearnBench\u002Fblob\u002Fmain\u002Fskills\u002Fhuman_authored\u002Fvideo-object-counting\u002Fobject_counter\u002FSKILL.md",[66],{"categoryNo":67,"categoryName":68,"taxonomyType":69,"taxonomyTypeDesc":70},"01JBSEEDCAT000000000000016","图像与多媒体","SCENARIO","使用场景",[72,29,73,74,75,76],"object counter","vision","computer","Computer Vision","image",{"skillNo":78,"name":79,"nameZh":6,"description":80,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":81,"categories":82,"keywords":86,"publishedAt":32},"01M6KM51TMVXG79MMPCS6Y8G86","travel-planning-workflow","Invoke this skill when performing travel planning. It defines a structured workflow for generating itineraries. Following this workflow is essential to ensure the final plan satisfies user requirements.","https:\u002F\u002Fgithub.com\u002Fcxcscmu\u002FSkillLearnBench\u002Fblob\u002Fmain\u002Fskills\u002Fhuman_authored\u002Ftravel-planning\u002Ftravel-planning-workflow\u002FSKILL.md",[83],{"categoryNo":84,"categoryName":85,"taxonomyType":69,"taxonomyTypeDesc":70},"01JBSEEDCAT000000000000023","Skill与智能体开发",[87,88,89,90,91,92,93,94,95,30,96,97,98,99,100],"workflow","planning","planning workflow","travel","travel planning","travel planning workflow","structured","performing","plan","user","it","requirements","essential","itineraries",{"skillNo":102,"name":103,"nameZh":6,"description":104,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":105,"categories":106,"keywords":107,"publishedAt":32},"01M6KM4DY820XHS9YTSK5QP5HD","bundled-dataset-search-tool","invoke this skill when you need to perform database search for travel planning. This skill provides some useful pre-packaged tools to look up accommodations, attractions, cities, driving distance, flights, and restaurants from the bundled dataset.","https:\u002F\u002Fgithub.com\u002Fcxcscmu\u002FSkillLearnBench\u002Fblob\u002Fmain\u002Fskills\u002Fhuman_authored\u002Ftravel-planning\u002Fbundled-dataset-search-tool\u002FSKILL.md",[],[108,109,110,111,112,113,114,90,115,116,91,117,30,118,88],"search","tool","dataset","bundled dataset search tool","restaurants","Tools","up","look","flights","you","database",{"skillNo":120,"name":121,"nameZh":6,"description":122,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":123,"categories":124,"keywords":125,"publishedAt":32},"01M6KM465BB9PQ0B548HQSVAWZ","glm-output","Output-processing guidance for GLM tasks. Especially useful after glm-basics and glm-calibration when you need verifier-matching metrics from output.nc and a final \u002Froot\u002Fmetrics.json.","https:\u002F\u002Fgithub.com\u002Fcxcscmu\u002FSkillLearnBench\u002Fblob\u002Fmain\u002Fskills\u002Fhuman_authored\u002Ftemperature-simulation\u002Fglm-output\u002FSKILL.md",[],[126,127,128,30,129,130,131,132,133,20,134,135,117,136,137],"glm output","output","glm","tasks","metrics","json","basics","guidance","verifier","processing","root","calibration",{"skillNo":139,"name":140,"nameZh":6,"description":141,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":142,"categories":143,"keywords":145,"publishedAt":32},"01M6KM40KY3QWDW5W97S39R2BC","glm-calibration","Calibration guidance for GLM tasks. Often most effective after glm-basics has clarified the setup; glm-output is the companion skill for exact final metric computation.","https:\u002F\u002Fgithub.com\u002Fcxcscmu\u002FSkillLearnBench\u002Fblob\u002Fmain\u002Fskills\u002Fhuman_authored\u002Ftemperature-simulation\u002Fglm-calibration\u002FSKILL.md",[144],{"categoryNo":84,"categoryName":85,"taxonomyType":69,"taxonomyTypeDesc":70},[146,137,128,127,133,132,129,147,148,30,149],"glm calibration","setup","companion","metric",{"skillNo":151,"name":152,"nameZh":6,"description":153,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":154,"categories":155,"keywords":156,"publishedAt":32},"01M6KM3VHHVSGMGHJQB3PPT6YG","glm-basics","A good starting point for GLM calibration tasks. Use to inspect glm3.nml, confirm how GLM runs, and identify the relevant files before moving on to calibration and output evaluation.","https:\u002F\u002Fgithub.com\u002Fcxcscmu\u002FSkillLearnBench\u002Fblob\u002Fmain\u002Fskills\u002Fhuman_authored\u002Ftemperature-simulation\u002Fglm-basics\u002FSKILL.md",[],[157,128,132,158,146,137,30,127,159,160,161,162,163,164,129],"glm basics","inspect","relevant","files","good","use","how","Evaluation"]