# weiyithu/SurroundOcc

[ICCV 2023] SurroundOcc: Multi-camera 3D Occupancy Prediction for Autonomous Driving

Repository: https://github.com/weiyithu/SurroundOcc
Canonical: https://ross.abutalabs.com/products/surroundocc
Language: Python
License: Apache-2.0
License Family: permissive
Topics: 3d-reconstruction, 3d-semantic-segmentation, occupancy, occupancy-prediction
Last push: 2025-06-20T12:47:59+00:00

## Health v2 (maintenance only)
Score: 43/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 27, release rhythm 35, longevity 91
- inputs: {"age_days": 1284, "days_push": 439, "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 1081, forks 136 (observed 2026-08-28T04:03:30.887041+00:00)

## What it is
SurroundOcc is the official PyTorch implementation of an ICCV 2023 paper predicting dense 3D volumetric occupancy from multi-camera images for autonomous driving. It also includes a pipeline that generates dense occupancy ground truth from sparse LiDAR points using Poisson Reconstruction without extra human annotation.

## Use cases
- predict 3d occupancy from multi-camera images
- generate dense occupancy ground truth from sparse lidar
- 3d semantic occupancy prediction for autonomous driving
- train a multi-camera 3d scene reconstruction model
- run occupancy prediction on private driving data
- evaluate pretrained 3d occupancy models on nuScenes

## When to choose
- you need state-of-the-art multi-camera 3D occupancy prediction for driving research
- you want to create dense occupancy labels from existing 3D detection and segmentation annotations
- you are benchmarking occupancy prediction on nuScenes or similar datasets

## When to avoid
- you need a production-ready perception stack rather than research code
- your project requires real-time occupancy prediction on embedded hardware out of the box
- you work outside autonomous driving or lack GPU resources for training

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, computer-vision, image-processing
- domain: autonomous-vehicles, computer-vision, deep-learning, artificial-intelligence
- platform: python
- tags: 3d-occupancy-prediction, multi-camera, 3d-semantic-segmentation, autonomous-driving, iccv-2023, lidar, nuscenes, research-code, linux, gpu

## Member repositories
- weiyithu/SurroundOcc (main) score 43

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:30.887041+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 2026-08-30T06:51:24.655760+00:00, confidence not recorded.
  - readme: https://github.com/weiyithu/SurroundOcc (fetched 2026-08-28T04:03:30.887041+00:00, sha 8bdd36e42d10)
- Data as of 2026-08-30T08:39:29.467469+00:00.