JL1-CD: Remote Sensing Change Detection Benchmark for Knowledge Distillation

JL1-CD is a remote sensing change detection benchmark created by circleLZY and updated in September 2025. The dataset provides paired satellite imagery and change labels specifically designed to support and evaluate a robust multi-teacher knowledge distillation framework.

Use Cases

• Training change detection models using the JL1-CD benchmark masks
• Evaluating multi-teacher knowledge distillation frameworks on remote sensing imagery
• Benchmarking pixel-level change classification performance in deep learning models

Strengths

• Designed specifically for multi-teacher knowledge distillation research
• Provides a benchmark for remote sensing change detection
• Updated as of late 2025

Limitations

• Missing explicit record counts and file size in the provided metadata
• Geographic scope and sensor specifications are not defined in the source
• License terms are not explicitly stated in the repository metadata

Provenance

Source

circleLZY via GitHub

Collection Method

Remote sensing imagery collection and annotation for the JL1-CD benchmark

Freshness

Last updated September 2025.