LiT Tuned Models for Efficient Species Detection

Andre Nakkab, Benjamin Feuer, Chinmay Hegde

Keywords: vision-language, AI, ecology, training, iNaturalist, contrastive loss, agriculture, LiT

TL;DR: We introduce a simple methodology for adapting any fine-grained image classification dataset for distributed pretraining of vision-language models, using iNaturalist 2021 as a primary example to simulate agricultural/ecological monitoring data.

Abstract: Recent advances in training vision-language models have demonstrated unprecedented robustness and transfer learning effectiveness; however, standard computer vision datasets are image-only, and therefore not well adapted to such training methods. Our paper introduces a simple methodology for adapting any fine-grained image classification dataset for distributed vision-language pretraining. We implement this methodology on the challenging iNaturalist-2021 dataset, comprised of approximately 2.7 million images of macro-organisms across 10,000 classes, and achieve a new state-of-the art model in terms of zero-shot classification accuracy. Somewhat surprisingly, our model (trained using a new method called locked-image text tuning) uses a pre-trained, frozen vision representation, proving that language alignment alone can attain strong transfer learning performance, even on fractious, long-tailed datasets. Our approach opens the door for utilizing high quality vision-language pretrained models in agriculturally relevant applications involving species detection.

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/lit-tuned-models-for-efficient-species/code)