# mpatacchiola/deepgaze

Computer Vision library for human-computer interaction. It implements Head Pose and Gaze Direction Estimation Using Convolutional Neural Networks, Skin Detection through Backprojection, Motion Detection and Tracking, Saliency Map.

Repository: https://github.com/mpatacchiola/deepgaze
Canonical: https://ross.abutalabs.com/products/deepgaze
Language: Python
License: MIT
License Family: permissive
Topics: convolutional-neural-networks, motion-tracking, color-detection, face-detection, skin-detection, motion-detection, head-pose-estimation, human-computer-interaction, histogram-comparison, histogram-intersection, cnn, particle-filter, saliency-map
Last push: 2024-05-09T20:37:09+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3830, "days_push": 846, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1880, forks 480 (observed 2026-08-28T04:05:48.569506+00:00)

## What it is
Deepgaze is a Python computer vision library for human-computer interaction built on OpenCV and TensorFlow. It provides CNN-based head pose and gaze direction estimation, face detection, skin/color detection, motion detection and tracking, and saliency map generation.

## Use cases
- estimate head pose from camera images with a CNN
- detect where a person is looking (gaze direction estimation)
- detect faces in video frames
- detect skin regions in images via histogram backprojection
- detect and track motion in video with a particle filter
- compute saliency maps to find visually salient regions
- classify images using histogram intersection

## When to choose
- you need head pose or gaze estimation in Python with OpenCV/TensorFlow
- you are building human-computer interaction or robotics attention systems
- you want classic CV techniques like skin detection, motion detection, and particle-filter tracking in one library

## When to avoid
- you need actively developed state-of-the-art models or GPU-optimized inference
- you need real-time performance on embedded hardware
- you are not working in Python or cannot use the pinned OpenCV/TensorFlow versions

## Facets
- artifact type: library
- maturity: maintenance
- function: computer-vision, image-processing, machine-learning, deep-learning
- domain: computer-vision, machine-learning, robotics
- platform: python, cross-platform
- tags: head-pose-estimation, gaze-estimation, face-detection, skin-detection, motion-detection, motion-tracking, saliency-map, particle-filter, opencv, tensorflow, cnn, human-computer-interaction

## Member repositories
- mpatacchiola/deepgaze (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:48.569506+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:14:01.102155+00:00, confidence not recorded.
  - readme: https://github.com/mpatacchiola/deepgaze (fetched 2026-08-28T04:05:48.569506+00:00, sha cecdb0fba8dc)
- Data as of 2026-08-30T08:39:29.467469+00:00.