---
id: imgviz
version: "2.1.0"
license: MIT
license_treatment: permissive
maintenance: active
---

# imgviz — Image Visualization Tools

License: permissive · Maintenance: active · Downloads: 103.4K/mo

## What it is and what it does

imgviz is a Python library for converting and visualizing non-RGB image data—such as depth maps, semantic segmentation labels, instance masks, and custom per-pixel flags—into displayable RGB images. It wraps common visualization patterns (colorization with colormaps, label-to-color mapping, instance bounding boxes and masks, pie-chart glyphs) and provides utilities for image composition and tiling. The library depends on numpy for array operations, pillow for image I/O and basic manipulation, and cmap for colormap support.

The package is designed for computer vision and machine learning workflows where intermediate representations (depth, class labels, instance IDs) need to be rendered for inspection, debugging, or publication. It handles the boilerplate of mapping scalar or categorical data to colors, overlaying annotations on images, and arranging multiple visualizations into a grid.

Use it for:

- Visualize depth sensor output by colorizing depth maps with a perceptually uniform colormap.
- Convert semantic segmentation label maps to RGB images with per-class colors and optional class name overlays.
- Render instance segmentation results by drawing bounding boxes, masks, and captions on a base image.
- Tile and arrange multiple visualization outputs (RGB, depth, labels, masks) into a single composite image for comparison.
- Overlay per-instance metadata (flags, statistics) as pie-chart glyphs at instance centroids.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

imgviz provides image visualization utilities for converting, colorizing, and compositing images—including depth maps, semantic labels, instance masks, and custom overlays—into displayable RGB outputs.

Yes. imgviz is actively maintained, has no known vulnerabilities, low install friction, and a permissive MIT license. Install it if you work with non-RGB image data (depth, labels, masks) in computer vision or ML and want a lightweight, ready-made toolkit for visualization rather than writing custom colorization and compositing code.

## Install

pip install imgviz
uv add imgviz
poetry add imgviz

## Installing imgviz

Before you install: Low friction: pure Python wheel, three lightweight runtime dependencies (numpy, pillow, cmap), and active maintenance with a release within the last 65 days.

License in practice: MIT license is permissive; you can use, modify, and distribute imgviz freely in commercial or private projects with minimal restrictions.

Quickstart:

pip install imgviz

import numpy as np
import imgviz

data = imgviz.data.arc2017()
rgb = data["rgb"]
gray = imgviz.rgb2gray(rgb)
depth = data["depth"]
depthviz = imgviz.colorize(depth, vmin=0.3, vmax=1)

Requires Python 3.10 or later.

Verify before relying:

- Whether optional dependencies (e.g., skimage) are required for specific visualization functions or only extend functionality.
- Performance characteristics when processing large images or batch operations.

## Package facts

- License: MIT (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 103.4K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

image visualization tools, colorize depth maps, label to rgb conversion, instance mask visualization, image compositing and tiling, computer-vision, visualization, image-processing

[View on SkillFed](https://skillfed.io/packages/imgviz) · [View on PyPI](https://pypi.org/project/imgviz/)