OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Models

DEEPLABV3

DEEPLABV3 is a Python and PyTorch implementation of DeepLabV3 trained on Cityscapes, with a pretrained checkpoint and documented workflows for model training, evaluation and visualization.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

DEEPLABV3 — OpenViST project profile in AI Models.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you need a code-based DeepLabV3 reference for Cityscapes training, validation, evaluation or segmentation visualization.

Who it is for

  • Computer-vision practitioners exploring semantic segmentation of street-scene imagery.
  • Researchers or developers studying a Cityscapes-based DeepLabV3 training and evaluation workflow.

Documented uses

  • Use the included pretrained model in the documented Cityscapes evaluation workflow.
  • Train a model using the repository’s Cityscapes workflow.
  • Generate and inspect predicted segmentation images.
  • Review a concrete PyTorch implementation of DeepLabV3.
Before you choose

Requirements & limits

The repository is publicly accessible on GitHub and lists an MIT license. The documented Cityscapes dataset download requires registration with Cityscapes.

  • The implementation is Python-based and uses PyTorch.
  • The supplied checkpoint is described as trained on Cityscapes and Berkeley DeepDrive imagery.
  • The README’s environment instructions specify Ubuntu 16.04, CUDA 9.1, NVIDIA Docker and PyTorch 0.4, which may require adaptation for other environments.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Prepare Cityscapes data for the repository

Set up the repository’s documented Cityscapes data layout for its preprocessing and model workflows.

Before you start

  • A Cityscapes account to download the dataset packages.
  • The repository cloned locally.
  • A shell environment with wget and unzip.

Steps

  1. Download the Cityscapes training, validation and test annotation package and left-image package using the README’s cookie-based download instructions.

  2. Unzip both packages.

  3. Create deeplabv3/data/cityscapes and move the extracted gtFine and leftImg8bit directories into it.

  4. From the repository, run python deeplabv3/utils/preprocess_data.py as directed before training or validation.

Expected outcome

The dataset is placed in the repository layout expected by the documented preprocessing workflow.

Guide limitations

  • Cityscapes registration is required for the documented dataset downloads.
  • The guide covers data preparation, not successful execution of training or evaluation.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
17 October 2026
Original website title
fregu856/deeplabv3 · Official homepage title · retained in its source language
How we review sourcesSuggest a correction