Using synthetic images generated from three real photos, contrastive fine-tuning of pretrained vision encoders yields personalized representations that beat the pretrained features on most classification, retrieval, detection, and segmentation evaluations.
This entailed looking through the training and test data and making sure that no test images were from the same sequential footage as the train data
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Personalized Representation from Personalized Generation
Using synthetic images generated from three real photos, contrastive fine-tuning of pretrained vision encoders yields personalized representations that beat the pretrained features on most classification, retrieval, detection, and segmentation evaluations.