ImageBind explores a shared embedding space that connects different kinds of data.
CLIPasso
CLIPasso explores semantically aware object sketching as a research project.
About CLIPasso
CLIPasso is a research project that converts input images of objects into abstract sketches while preserving key semantic and structural features. It supports varying levels of abstraction controlled by the number of strokes used, and does not require explicit sketch datasets for training, as it relies on the CLIP image encoder for semantic and geometric guidance. The method works across diverse object categories without category-specific training constraints.
Key features
- Converts input object images to abstract sketches
- Adjustable abstraction via variable stroke count
- CLIP-based semantic and geometric sketch guidance
- No explicit sketch dataset required for training
- Supports SVG brush style editing
- Generalizes across diverse object categories
Best for: Researchers and developers in computer graphics and generative AI focused on sketch synthesis and semantic image abstraction.
Good for
- No reliance on limited sketch training datasets
- Maintains recognizability even with minimal stroke counts
- Generalizes to diverse object categories without category-specific training
CLIPasso at a glance
- Website
- clipasso.github.io/clipasso
- Category
- AI & Machine Learning › Models & ML Research
- Type
- Website
- Pricing
- Not confirmed. Check the site
- Platforms
- Web
- Open source
- Not stated
- Last checked
- Oct 6, 2026 · Online when checked
- Listed
- Oct 5, 2026
CLIPasso alternatives
Tools and sites you could use instead, ranked by how closely they match.
Use SmolVLA, a vision-language-action model trained on data from the LeRobot community, for robotics research.
The Stanford Alpaca project presents a replicable instruction-following language model and its research materials.
This research paper presents Word-As-Image, a method for incorporating semantic meaning into typography.
Google Research introduced Imagen as a text-to-image diffusion model in this research paper.
Falcon LLM is described as a generative language model from the Technology Innovation Institute for a range of AI applications.
Popular in Models & ML Research
Well-known listings from the same topic.
ERNIEernie.baidu.comTokenVersetoken-verse.github.iowTAObittensor.comStable Diffusionhuggingface.co/CompVis/stable-diffusion-v1-4AI Indexhai.stanford.edu/ai-index
CLIPasso FAQ
How does CLIPasso generate object sketches?
It optimizes Bézier curve stroke parameters using a CLIP-based perceptual loss and differentiable rasterizer.
Does CLIPasso require sketch datasets for training?
No, it uses a pretrained CLIP model for guidance and does not rely on explicit sketch training data.
What controls the abstraction level of generated sketches?
The number of strokes used to define the sketch determines the level of abstraction.
Is something wrong or out of date?
Tell us and an editor will check it.
Own this website?
Claim this listing
Claim the listing and suggest updates yourself. An editor checks every claim first.