Trashnet: Garbage Image Classification for Torch-Based CNNs

Gary Thung released this image collection for garbage classification, with the latest update provided in June 2023. It provides visual data specifically for training Torch-based convolutional neural networks to identify different types of trash.

Use Cases

• Training image classification models to recognize garbage categories
• Benchmarking deep learning architectures on waste sorting tasks
• Developing automated recycling systems using the labeled trash images

Strengths

• MIT license allows for flexible academic and commercial reuse
• Domain-specific focus on waste management and recycling
• Includes associated Torch-based CNN implementation for immediate model training

Limitations

• Total record count and class distribution are not explicitly defined in the source metadata
• Lack of geographic or environmental metadata for the captured images

Provenance

Source

garythung via GitHub

Freshness

Last updated June 2023.