CVAIDec 31, 2023

SynCDR : Training Cross Domain Retrieval Models with Synthetic Data

arXiv:2401.00420v22 citationsh-index: 7Has Code
AI Analysis

This addresses a practical challenge in cross-domain retrieval for scenarios like biometric sensors or sketch-to-image search, where training data is limited, but it is incremental as it builds on existing translation and retrieval methods.

The paper tackles the problem of cross-domain retrieval when training data lacks shared categories across domains, by generating synthetic data via category-preserving translation, and finds that using large-scale pre-trained text-to-image diffusion models for this translation leads to more accurate models, outperforming prior art by up to 15%.

In cross-domain retrieval, a model is required to identify images from the same semantic category across two visual domains. For instance, given a sketch of an object, a model needs to retrieve a real image of it from an online store's catalog. A standard approach for such a problem is learning a feature space of images where Euclidean distances reflect similarity. Even without human annotations, which may be expensive to acquire, prior methods function reasonably well using unlabeled images for training. Our problem constraint takes this further to scenarios where the two domains do not necessarily share any common categories in training data. This can occur when the two domains in question come from different versions of some biometric sensor recording identities of different people. We posit a simple solution, which is to generate synthetic data to fill in these missing category examples across domains. This, we do via category preserving translation of images from one visual domain to another. We compare approaches specifically trained for this translation for a pair of domains, as well as those that can use large-scale pre-trained text-to-image diffusion models via prompts, and find that the latter can generate better replacement synthetic data, leading to more accurate cross-domain retrieval models. Our best SynCDR model can outperform prior art by up to 15\%. Code for our work is available at https://github.com/samarth4149/SynCDR .

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes