Classify images and match them with text using CLIP vision-language similarities.
npx skills add https://github.com/JKhyro/HERMES-AGENT --skill clip-jkhyro
Please help me install this Agent Skill. Skill: clip Source: https://github.com/JKhyro/HERMES-AGENT/tree/main/skills/mlops/models/clip Command: npx skills add https://github.com/JKhyro/HERMES-AGENT --skill clip-jkhyro
This Skill helps you connect images and text so you can classify visual content, search by meaning, and compare captions without building a labeled dataset.
Use the clip skill to classify the attached image against a short list of labels and return the most likely match with confidence.
High-intent search queries and answers about installing and using this skill.
Zero-shot image classification matches images to natural-language prompts by computing CLIP vision-language similarities. You provide an image and a list of text labels, and it ranks the best match using normalized embeddings and cosine-similarity scoring.
Semantic image search matches text queries to images using cross-modal retrieval. It processes batches by loading a transformer-based CLIP model, preprocessing images, generating normalized embeddings, and ranking cosine-similarity scores.
Cross-modal retrieval connects images and text by computing CLIP vision-language similarities. It generates normalized embeddings for both formats and ranks their cosine-similarity scores to find matching content across modalities without labeled datasets.
You do not need a labeled dataset for content moderation or similarity checks using CLIP. It performs zero-shot inference by comparing normalized embeddings and ranking cosine-similarity scores across single images or batches.
Deduplicating a photo library by visual meaning uses CLIP to compute vision-language similarities. It preprocesses batch images into normalized embeddings and ranks cosine-similarity scores to identify matching content without training a custom model.
HarriesConnects images and text in a shared embedding space using CLIP.
zerwizClassifies images and retrieves cross-modal matches using natural-language prompts and cosine similarity scoring without training data.
eddieluengCompute image and text embeddings for zero-shot classification and retrieval.
photonics-dhlCompute image-text similarity scores with pretrained CLIP models.
