# google-research-datasets/Objectron

Objectron is a dataset of short, object-centric video clips. In addition, the videos also contain AR session metadata including camera poses, sparse point-clouds and planes. In each video, the camera moves around and above the object and captures it from different views. Each object is annotated with a 3D bounding box. The 3D bounding box describes the object’s position, orientation, and dimensions. The dataset contains about 15K annotated video clips and 4M annotated images in the following categories: bikes, books, bottles, cameras, cereal boxes, chairs, cups, laptops, and shoes

Repository: https://github.com/google-research-datasets/Objectron
Canonical: https://ross.abutalabs.com/products/objectron
Language: Jupyter Notebook
License: NOASSERTION
License Family: other
Topics: deep-learning, computer-vision, machine-learning, python, tensorflow, pytorch, 3d-vision, 3d-reconstruction, ai, 3d, neural-network, dataset, augmented-reality
Last push: 2026-03-06T17:27:37+00:00

## Health v2 (maintenance only)
Score: 54/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 70, release rhythm 8, longevity 100
- inputs: {"age_days": 2144, "days_push": 180, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2349, forks 266 (observed 2026-08-28T04:06:40.175765+00:00)

## What it is
Objectron is a dataset of ~15K object-centric video clips with over 4M annotated images, including 3D bounding box annotations, camera poses, sparse point-clouds, and plane metadata. It ships with parsing tools and tutorials for TensorFlow and PyTorch, covering nine object categories such as bikes, books, bottles, cameras, chairs, cups, laptops, and shoes.

## Use cases
- train 3D object detection models from video
- get 3D bounding box annotations for everyday objects
- research multi-view object pose estimation
- benchmark 3D IoU metrics for oriented bounding boxes
- build augmented reality object tracking features
- download annotated AR session data with camera poses and point clouds

## When to choose
- you need labeled 3D bounding boxes for object detection research
- you want multi-view video data with AR metadata like camera poses and planes
- you need ready-made tf.record pipelines for TensorFlow or PyTorch

## When to avoid
- you need 2D-only image classification labels
- you require objects outside the nine supported categories
- you need real-time inference rather than training data

## Facets
- artifact type: dataset
- maturity: stable
- function: machine-learning, computer-vision, data-science
- domain: computer-vision, machine-learning, deep-learning, artificial-intelligence
- platform: python, cross-platform
- tags: 3d-object-detection, 3d-bounding-boxes, augmented-reality, video-dataset, pose-estimation, tensorflow, pytorch, 3d-vision

## Member repositories
- google-research-datasets/Objectron (main) score 54

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:40.175765+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 2026-08-30T02:36:57.752556+00:00, confidence not recorded.
  - readme: https://github.com/google-research-datasets/Objectron (fetched 2026-08-28T04:06:40.175765+00:00, sha c0f02fea7dae)
- Data as of 2026-08-30T08:39:29.467469+00:00.