Datasets:
The dataset viewer is not available for this dataset.
Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
EditCLIP: Representation Learning for Image Editing
π Introduction
The TOP-Bench-X dataset offers Query and Exemplar image pairs tailored for exemplar-based image editing. We built it by adapting the TOP-Bench dataset from InstructBrush (also uploaded huggingface version at Aleksandar/InstructBrush-Bench). Specifically, we use the original training split to generate exemplar images and the test split to supply their corresponding queries. In total, TOP-Bench-X comprises 1,277 samples, including 257 distinct exemplars and 124 unique queries.
π‘ Abstract
We introduce EditCLIP, a novel representation-learning approach for image editing. Our method learns a unified representation of edits by jointly encoding an input image and its edited counterpart, effectively capturing their transformation. To evaluate its effectiveness, we employ EditCLIP to solve two tasks: exemplar-based image editing and automated edit evaluation. In exemplar-based image editing, we replace text-based instructions in InstructPix2Pix with EditCLIP embeddings computed from a reference exemplar image pair. Experiments demonstrate that our approach outperforms state-of-the-art methods while being more efficient and versatile. For automated evaluation, EditCLIP assesses image edits by measuring the similarity between the EditCLIP embedding of a given image pair and either a textual editing instruction or the EditCLIP embedding of another reference image pair. Experiments show that EditCLIP aligns more closely with human judgments than existing CLIP-based metrics, providing a reliable measure of edit quality and structural preservation.
π§ Data explained
Each sample consists of 4 images (2 pairs of images) and metadata, specifically:
- input_test β the query image (I_q) from the test split (βbeforeβ image you want to edit)
- input_gt β the ground-truth edited version of that query image (βafterβ image for the test)
- exemplar_input β the exemplarβs input image (I_i) from the training split (βbeforeβ image of the exemplar)
- exemplar_edit β the exemplarβs edited image (I_e) from the training split (βafterβ image of the exemplar)
π Citation
@article{wang2025editclip,
title={EditCLIP: Representation Learning for Image Editing},
author={Wang, Qian and Cvejic, Aleksandar and Eldesokey, Abdelrahman and Wonka, Peter},
journal={arXiv preprint arXiv:2503.20318},
year={2025}
}
π³ License
This dataset is mainly a variation of TOP-Bench, confirm the license from the original authors.
- Downloads last month
- 40