The Curious Layperson: Fine-Grained Image Recognition Without Expert Labels

Abstract

Most of us are not experts in specific fields, such as ornithology. Nonetheless, we do have general image and language understanding capabilities that we use to match what we see to expert resources. This allows us to expand our knowledge and perform novel tasks without ad-hoc external supervision. On the contrary, machines have a much harder time consulting expert-curated knowledge bases unless trained specifically with that knowledge in mind. Thus, in this paper we consider a new problem: fine-grained image recognition without expert annotations, which we address by leveraging the vast knowledge available in web encyclopedias. First, we learn a model to describe the visual appearance of objects using non-expert image descriptions. We then train a fine-grained textual similarity model that matches image descriptions with documents on a sentence-level basis. We evaluate the method on two datasets (CUB-200 and Oxford-102 Flowers) and compare with several strong baselines and the state of the art in cross-modal retrieval. Code is available at: https://github.com/subhc/clever.

ILSGAN: Independent Layer Synthesis for Unsupervised Foreground-Background Segmentation

“…On another front, methods based on clustering attain unsupervised segmentation through self-supervised learning with mutual information maximization (Ji, Henriques, and Vedaldi 2019;Ouali, Hudelot, and Tami 2020) or contrastive learning (Cho et al 2021;Van Gansbeke et al 2021;Choudhury et al 2021;Hwang et al 2019). However, when these methods are applied to binary segmentation of foreground and background, they show disadvantages (Benny and Wolf 2020) in segmentation performance compared to methods based on layered GANs.…”