Transparent Fragment Images and Masks with HDRI Lighting

TransFrag27K is a large-scale dataset containing 27,000 images and masks at 640×480 resolution. It covers fragments of common everyday glassware, incorporating over 150 background textures and 100 HDRI environment lightings. The dataset was created by chenbr7 and last updated in August 2025.

Use Cases

• Train segmentation models to predict pixel-level masks from the provided image data.
• Evaluate model robustness using the dataset's 150 varied background textures.
• Study material property rendering with the 100 HDRI environment lightings included.
• Develop algorithms for handling refractive and transmissive properties of transparent objects like glassware.

Strengths

• 27,000 image and mask pairs provide a substantial sample size.
• High-resolution images at 640×480 pixels offer detailed visual data.
• Includes over 150 background textures and 100 HDRI lightings for diversity.

Limitations

• Dataset focuses solely on fragments of glassware, limiting object variety.
• All data is synthetic, which may not fully capture real-world conditions.
• Specific details on class distribution or label accuracy are unknown.

Provenance

Source

chenbr7 on Hugging Face.

Collection Method

Synthetic generation incorporating textures and HDRI lighting.

Freshness

Last updated on 2025-08-19.