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

Tencent Ml Images

Tencent ML-Images is an open-source computer-vision resource containing annotated training and validation image references from ImageNet and Open Images, plus a ResNet-101 pretrained on the dataset. Its labels cover up to 11,166 categories, and the repository documents workflows for pretraining, fine-tuning, single-label classification and feature extraction.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Tencent Ml Images — 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 a computer-vision project needs broad multi-label annotations or a pretrained ResNet-101 as a research starting point, and its legacy environment requirements fit the intended workflow.

Who it is for

  • Computer-vision researchers working with multi-label image data.
  • Developers prototyping image classifiers or adapting a pretrained ResNet-101.
  • Practitioners evaluating image models against the provided validation references.

Documented uses

  • Train multi-label image classifiers using the annotations.
  • Evaluate image-recognition models with the validation set.
  • Start transfer learning from the pretrained ResNet-101.
  • Build image-category prediction prototypes using the dataset labels.
Before you choose

Requirements & limits

  • The dataset provides image IDs and URLs rather than the image files themselves; ImageNet images must be obtained from ImageNet.
  • The documented environment is Linux, Python 2.7 and TensorFlow 1.6 or later.
  • The repository reports 80.73% top-1 accuracy on ImageNet via transfer learning; this is a reported result, not a performance promise for other uses.
  • The README describes image references from both ImageNet and Open Images, with annotations including class indices and confidence values.

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

Obtain ImageNet images referenced by ML-Images

Use the project’s ImageNet IDs to select the corresponding training or validation images.

Before you start

  • Access to the ImageNet database.
  • The repository’s train_image_id_from_imagenet.txt and val_image_id_from_imagenet.txt files.

Steps

  1. Download the ImageNet database using its documented image-download route.

  2. Use the training or validation ID list to extract the corresponding images.

Expected outcome

Image files selected from ImageNet to match the project’s listed IDs and annotations.

Guide limitations

  • The repository provides IDs rather than the original image files.
  • The README notes that some images may be repeated under different IDs.
  • ImageNet URLs are described as having expired, so the README directs users to obtain the database rather than rely on those URLs.
Documentation 1 ↗
Evidence and transparency

Sources & review

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