On zero-shot recognition of generic objects
This work identifies critical flaws in a widely used benchmark for zero-shot object recognition, which could impact researchers and practitioners in computer vision.
The paper argues that the poor performance of zero-shot learning models on the Imagenet benchmark is due to structural flaws and bias, showing that actual accuracy is higher than previously thought, and proposes a new benchmark to address these issues.
Many recent advances in computer vision are the result of a healthy competition among researchers on high quality, task-specific, benchmarks. After a decade of active research, zero-shot learning (ZSL) models accuracy on the Imagenet benchmark remains far too low to be considered for practical object recognition applications. In this paper, we argue that the main reason behind this apparent lack of progress is the poor quality of this benchmark. We highlight major structural flaws of the current benchmark and analyze different factors impacting the accuracy of ZSL models. We show that the actual classification accuracy of existing ZSL models is significantly higher than was previously thought as we account for these flaws. We then introduce the notion of structural bias specific to ZSL datasets. We discuss how the presence of this new form of bias allows for a trivial solution to the standard benchmark and conclude on the need for a new benchmark. We then detail the semi-automated construction of a new benchmark to address these flaws.