ReEdit-Bench: Benchmark Dataset for Exemplar-Based Image Editing

ReEdit-Bench is a curated dataset of approximately 1,500 samples for evaluating exemplar-based image editing methods. It was created by tarun-menta and presented in a WACV '25 paper. Each sample contains four images representing an exemplar edit pair.

Use Cases

• Benchmarking the performance of exemplar-based image editing models based on the provided image pairs.
• Training or fine-tuning diffusion models for image editing tasks based on the exemplar edit structure.
• Conducting research on multimodal image editing techniques based on the paired image samples.

Strengths

• Curated dataset of ~1,500 samples, providing a focused evaluation set.
• Structured with four images per sample, defining a clear exemplar edit operation.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Data may reflect bias inherent to the source and curation method.

Provenance

Source

huggingface

Collection Method

Curated dataset presented in an academic paper.

Freshness

Last updated 2025-06-02 19:06:41; freshness should be verified.