agentsclimarketplace

Run3 count objects optimized

Skill cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-gemini-3-flash-preview/video-object-counting/run3_count_objects_optimized

Counts occurrences of specific objects (coins, enemies, turtles) in a frame using a template matching script with optimized parameters.From its SKILL.md

Install
npx -y skills add cxcscmu/SkillLearnBench --skill run3_count_objects_optimized

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

SKILL.md

1.1 KB, 214 tokens by cl100k_base, as published. Nobody here has run it

To count objects accurately, call the count_objects_optimized.py script using python3. You must specify the counting tool mode and maintain a high confidence threshold to ensure fidelity.

Command structure: python3 count_objects_optimized.py --tool count --input_image <frame_path> --object_image <template_path> --dedup_min_dist <distance> --threshold 0.9

Arguments:

  • --tool count: Required parameter to activate the counting logic.
  • --input_image: The path to the keyframe being analyzed (e.g., /root/keyframes_001.png).
  • --object_image: The path to the template image (e.g., /root/coin.png).
  • --dedup_min_dist: The minimum pixel distance between detected objects to prevent double-counting the same object (e.g., 10 or 15 depending on object size).
  • --threshold 0.9: Sets the confidence threshold for a match to 0.9 to ensure only high-confidence detections are counted.

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.