covecto 0.1.0

Dual-engine image vectorization CLI
covecto-0.1.0 is not a library.

Covecto

Dual-engine image vectorization — pixel-exact for icons, smooth curves for art.

CI crates.io License: MIT Docker

Why Covecto?

Most vectorizers are one-trick — good at photos or good at icons, never both. Covecto ships two engines and auto-selects the right one:

Engine Best For Algorithm
PixelExact Icons, pixel art, screenshots Contiguous region flood-fill → boundary tracing → rectilinear SVG
Spline Photos, illustrations, artwork Color quantization → contour tracing → Bézier spline fitting
Auto (default) Everything Heuristic selection based on image characteristics

Install

# Cargo
cargo install covecto

# Docker
docker pull ghcr.io/codecoradev/covecto:latest

# Pre-built binary (Linux/macOS/Windows)
# → See https://github.com/codecoradev/covecto/releases

Quick Start

# Auto-detect best engine
covecto input.png -o output.svg

# Force pixel-exact (great for icons)
covecto icon.png --engine pixel-exact -o icon.svg

# Spline with photo preset
covecto photo.jpg --engine spline --preset photo --optimize -o photo.svg

# Use a built-in profile (tunes multiple params)
covecto logo.png --profile logo -o logo.svg

# Batch: entire directory
covecto ./icons/ --engine pixel-exact -o ./output/

# Batch: glob pattern
covecto "screenshots/*.png" -o ./vectorized/

# Output formats
covecto input.png --format pdf -o input.pdf
covecto input.png --format eps -o input.eps

CLI Reference

Global

covecto [COMMAND]

Commands:
  vectorize  Vectorize one or more images to SVG/PDF/EPS
  serve      Start HTTP API server

Options:
  -V, --version  Print version
  -h, --help     Print help

Vectorize

covecto vectorize [OPTIONS] <INPUT>
Flag Description Default
<INPUT> File, directory, or glob pattern —
-o, --output Output file (single) or directory (batch) stdout
--output-dir Output directory (batch alt) —
--output-template Filename template: {stem}, {name}, {ext} {stem}.{ext}
-F, --format Output format: svg, pdf, eps svg
-e, --engine Engine: auto, spline, pixel-exact auto
--preset vtracer preset: bw, poster, photo —
--profile Profile: icon, logo, photo, lineart —
--color-precision Color quantization precision (1–32) —
--filter-speckle Filter speckle noise threshold —
--corner-threshold Corner detection (0–180) —
--splice-threshold Path splice threshold (0–100) —
--color-mode color or binary —
--hierarchical stacked or cutout —
--path-simplify spline, polygon, or none —
--optimize Run SVG optimization false
--optimize-preset default, safe, or none default
--multipass Multiple optimization passes false
-R, --recursive Recurse into subdirectories false
--json JSON output to stdout false
--json-pretty Pretty-printed JSON false
--no-progress Disable progress bar false
--dry-run Preview without processing false

Profiles

Profile Engine Color Use Case
icon spline 4 colors Small icons, favicons
logo spline 8 colors, cutout Logos with transparency
photo spline 10 colors, stacked Photographs
lineart spline binary, cutout Line drawings, sketches

Serve

covecto serve --port 3000

HTTP API

Start the server, then send requests:

# Vectorize
curl -F "file=@input.png" -F "engine=auto" http://localhost:3000/v1/vectorize

# Vectorize to PDF
curl -F "file=@photo.jpg" -F "format=pdf" http://localhost:3000/v1/vectorize

# Optimize existing SVG
curl -F "file=@input.svg" -F "preset=safe" http://localhost:3000/v1/optimize

# Health check
curl http://localhost:3000/v1/health

# Metrics
curl http://localhost:3000/v1/metrics

Full API spec: openapi.yaml

Docker

# Pull
docker pull ghcr.io/codecoradev/covecto:latest

# Run with docker compose
docker compose up -d

# Or directly
docker run -p 3000:3000 ghcr.io/codecoradev/covecto:latest

# Vectorize a local file
docker run -v $(pwd):/data ghcr.io/codecoradev/covecto:latest \
  vectorize /data/input.png -o /data/output.svg

Rust Library

use covecto_core::{vectorize, VectorizeRequest, Engine, VectorizeConfig};
use image::RgbaImage;

let img = image::open("input.png").unwrap().to_rgba8();
let request = VectorizeRequest::new(img)
    .with_config(VectorizeConfig {
        engine: Engine::Auto,
        ..Default::default()
    });

let result = vectorize(&request).unwrap();
println!("{} paths in {}ms", result.metadata.path_count, result.metadata.processing_time_ms);

Documentation

Full docs at docs.covecto.dev

License

MIT