datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
Japanese-Materials
仓库信息
电报地址：https://t.me/vomebook ，有问题请在：https://huggingface.co/datasets/VoiceOfML/Japanese-Materials/discussions 提出。
此仓库存储日共资料：https://huggingface.co/datasets/VoiceOfML/Japanese-Materials/tree/main 。
请使用：https://voiceofml-search.hf.space/Japanese-Materials 进行文件检索（备用搜索站：https://vomebook.github.io/search/#/Japanese-Materials ）。
可使用：https://voiceofml-search.hf.space/Japanese-Materials?wide=1 进行仓库内容查看（备用站：https://voiceofml-search.hf.space/Japanese-Materials?wide=1 ）。… See the full description on the dataset page: https://huggingface.co/datasets/VoiceOfML/Japanese-Materials.apple-dms-materials
Apple Dense Material Segmentation (DMS) Dataset
A pixel-level material segmentation dataset containing ~41K images with dense annotations across 57 material categories. Originally released by Apple as part of the Dense Material Segmentation (DMS) research project.
Note: This is a mirror prepared for direct use with the HuggingFace 🤗 datasets library. The source images originate from Open Images V7, and material annotations were created by Apple. Some images (~6%) from the original… See the full description on the dataset page: https://huggingface.co/datasets/AllanK24/apple-dms-materials.pbr-materials
PBR Materials (MonoRelief V2 core)
Auto-generated PBR material set (10 materials). Each subfolder contains a complete PBR set:
Map
File
Format
Albedo
*_albedo.png
8-bit RGB
Height
*_height.png
16-bit grayscale
Normal
*_normal.png
8-bit RGB (OpenGL, Godot 4)
Roughness
*_roughness.png
8-bit grayscale
Metallic
*_metallic.png
8-bit grayscale
Pipeline
Generated using MonoRelief V2 (Fig. 7), Marigold-only variant:
Base depth — Marigold Depth… See the full description on the dataset page: https://huggingface.co/datasets/Zakhar1746/pbr-materials.verbosity-materials
Verbosity Research Materials
公开研究材料库：7篇论文、8个任务（4个website、4个slides）、40份作品（8份作者作品、32份AI作品）。
开始阅读 / Start here
下载完整 v1 ZIP（约288 MiB）
浏览展开目录
中文入门及离线限制
作品与模型来源清单
论文及作者原始链接
解压后打开 verbosity-materials-v1/index.html 浏览。HF文件仓库不作为这些HTML的运行网站。
Slides包含20份PDF，AI作品另附可编辑HTML和本地资源；作者slides仅有PDF。
论文正文是生成时使用的快照，完整论文通过原始链接访问。
Provenance and use
The v1 snapshot comes from run materials-share-20260918a. All 40 artifacts retain
provenance; the bundle includes SHA-256… See the full description on the dataset page: https://huggingface.co/datasets/Penguin-N/verbosity-materials.apple-dms-materials-v2
Apple Dense Material Segmentation (DMS) – Stratified 80/10/10 Split
A pixel-level material segmentation dataset containing ~41K images with dense annotations across 57 material categories. Originally released by Apple as part of the Dense Material Segmentation (DMS) research project.
This version uses a custom stratified 80/10/10 split (vs Apple's original 54/23/23) to maximise training data while maintaining representative validation and test sets.
Why a Custom Split?… See the full description on the dataset page: https://huggingface.co/datasets/AllanK24/apple-dms-materials-v2.materials-figure-qa
Materials Figure QA
This repository contains 250 figure-grounded multimodal question-answer pairs from recent materials-related arXiv papers. Each item requires interpreting a rendered figure and applying materials-science reasoning. Figures with aspect ratio greater than 4:1 were removed; retained figures were downsampled so their longest dimension is at most 2048 pixels.
The filtered split preserves the original deterministic split:
validation: 128 examples
test: 122 examples… See the full description on the dataset page: https://huggingface.co/datasets/gneubig/materials-figure-qa.cgaxis-pbr-materials-sample
CGAxis PBR Materials - Free Sample (Brick, Stone & Concrete)
A free, licensed sample of human-authored PBR material sets from CGAxis, a 3D content studio operating since 2008. This sample is a taster of the full CGAxis AI Data corpus (7,794 PBR material sets + 4,390 3D models) available for commercial AI-training licenses.
Every material ships as a complete, consistently named set of maps with explicit color space, bit depth and real-world scale, plus per-set metadata.json with… See the full description on the dataset page: https://huggingface.co/datasets/CGAxis/cgaxis-pbr-materials-sample.materialsmaterials_under_microscopepbr-materials_V2
PBR Materials (MonoRelief V2 core)
Auto-generated PBR material set (10 materials). Each subfolder contains a complete PBR set:
Map
File
Format
Albedo
*_albedo.png
8-bit RGB
Height
*_height.png
16-bit grayscale
Normal
*_normal.png
8-bit RGB (OpenGL, Godot 4)
Roughness
*_roughness.png
8-bit grayscale
Metallic
*_metallic.png
8-bit grayscale
Pipeline
Generated using MonoRelief V2 (Fig. 7), Marigold-only variant:
Base depth — Marigold Depth… See the full description on the dataset page: https://huggingface.co/datasets/Zakhar1746/pbr-materials_V2.Elf_encoded_flat_band_materialsPH_Gr11-Learning_Materialsa23d-pbr-materials-sample
A23D PBR Materials — Sample for AI/ML Training
A sample from the A23D corpus of 100,000+ human-authored PBR materials, spanning a wide range of real-world material categories. This sample is provided for AI teams to inspect the quality, consistency, and per-map ground truth of A23D materials before licensing the full corpus.
Every material is authored in-house to a single production standard — giving consistency across resolution, channel packing, map conventions, naming, and… See the full description on the dataset page: https://huggingface.co/datasets/A23D/a23d-pbr-materials-sample.fmd_materials
Dataset Card for "fmd_materials"
class_label:
names:
'0': fabric
'1': foliage
'2': glass
'3': leather
'4': metal
'5': paper
'6': plastic
'7': stone
'8': water
'9': wood
Source: https://people.csail.mit.edu/lavanya/fmd.html
normal_map_everyday_materialsMaterials-Workbench
Materials Workbench — receipts & artefacts
Canonical code/control: github.com/Zer0pa/Materials-Workbench
This dataset is the durability mirror for large/binary artefacts produced by the Zer0pa Materials Workbench lane. Code, control documents, and small text artefacts live on GitHub. PNGs, render outputs, recovery snapshots, and any future GPU-evidence packets are stored here.
What lives here
Path
What it is
receipts/product-page/
Render-time screenshots +… See the full description on the dataset page: https://huggingface.co/datasets/Zer0pa/Materials-Workbench.MaterialScope
MaterialScope
A manually annotated object detection dataset for compound material image to sub-image region detection, in COCO format.All 2,855 images are fully annotated — no unannotated images in the split.
Dataset Summary
Metric
Value
Total images
2,855
Total annotations
13,008
Categories
23
Avg. annotations per image
4
Max annotations per image
21
Annotation format
COCO JSON
Image format
JPG
Category Distribution… See the full description on the dataset page: https://huggingface.co/datasets/CMEG-IITR/MaterialScope.Cephalo-Wikipedia-MaterialsThis dataset is used to train the Cephalo models. Cephalo is a series of multimodal materials science focused vision large language models (V-LLMs) designed to integrate visual and linguistic data for advanced understanding and interaction in human-AI or multi-agent AI frameworks.
A novel aspect of Cephalo's development is the innovative dataset generation method. The extraction process employs advanced algorithms to accurately detect and separate images and their corresponding textual… See the full description on the dataset page: https://huggingface.co/datasets/lamm-mit/Cephalo-Wikipedia-Materials.materials-blip
Dataset Card for "materials-blip"
More Information needed
draineye-materials
DrainEye — Drainage Material Classification Dataset
An image classification dataset for identifying pipe and drainage materials,
assembled from public Roboflow and MINC-2500 sources.
Classes
asbestos — asbestos cement pipes and surfaces
hdpe — high-density polyethylene (HDPE/ПНД) plastic pipes
ceramic — ceramic surfaces and pipes
stone — stone and slag-like surfaces
Dataset Structure
Split
hdpe
ceramic
stone
asbestos
Total
train
384
1750
1750
1200… See the full description on the dataset page: https://huggingface.co/datasets/draineye/draineye-materials.waste-wizard-materials-listdreambooth_materials