NFT1000: A Visual Text Dataset For Non-Fungible Token Retrieval
CoRR(2024)
Abstract
With the rise of 'Metaverse' and 'Web3.0', NFT ( Non-Fungible Token ) has
emerged as a kind of pivotal digital asset, garnering significant attention. By
the end of November 2023, more than 1.4 billion NFT tokens have been minted
across various blockchain platforms. To effectively locate a satisfactory NFT
token, conducting searches within the extensive array of NFT data is essential.
The challenge in NFT retrieval is heightened due to the high degree of
similarity among different NFT tokens, in terms of regional and semantic
aspects. Achieving accurate and efficient retrieval within the large-scale,
highly similar NFT data presents a formidable challenge for both the academic
and industrial communities. In this paper, we will introduce a dataset named
'NFT Top1000 Visual Text Dataset'(henceforth, NFT1000), containing 7.56 million
image-text pairs, and being collected from 1000 most famous PFP NFT collections
by sales volume on the Ethereum blockchain. Based on the dataset, we test the
CLIP (Contrastive Language-Image Pretraining) models as a baseline.
Additionally, we also propose a concept of Comprehensive Variance Index (CVI in
short), which is a robust metric designed to assess the similarity and
retrieval difficulty of visual-text pairs data.
MoreTranslated text
AI Read Science
Must-Reading Tree
Example
![](https://originalfileserver.aminer.cn/sys/aminer/pubs/mrt_preview.jpeg)
Generate MRT to find the research sequence of this paper
Chat Paper
Summary is being generated by the instructions you defined