ELEVATER: A Benchmark and Toolkit for Evaluating Language-Augmented Visual Models
This provides a standardized evaluation platform for researchers in computer vision, addressing a gap in benchmarking tools for language-augmented models, though it is incremental as it builds on existing model evaluation concepts.
The paper tackles the challenge of evaluating the transferability of language-augmented visual models by introducing ELEVATER, a benchmark and toolkit that includes 20 image classification and 35 object detection datasets with external knowledge, along with automated hyper-parameter tuning and metrics for sample- and parameter-efficiency.
Learning visual representations from natural language supervision has recently shown great promise in a number of pioneering works. In general, these language-augmented visual models demonstrate strong transferability to a variety of datasets and tasks. However, it remains challenging to evaluate the transferablity of these models due to the lack of easy-to-use evaluation toolkits and public benchmarks. To tackle this, we build ELEVATER (Evaluation of Language-augmented Visual Task-level Transfer), the first benchmark and toolkit for evaluating(pre-trained) language-augmented visual models. ELEVATER is composed of three components. (i) Datasets. As downstream evaluation suites, it consists of 20 image classification datasets and 35 object detection datasets, each of which is augmented with external knowledge. (ii) Toolkit. An automatic hyper-parameter tuning toolkit is developed to facilitate model evaluation on downstream tasks. (iii) Metrics. A variety of evaluation metrics are used to measure sample-efficiency (zero-shot and few-shot) and parameter-efficiency (linear probing and full model fine-tuning). ELEVATER is a platform for Computer Vision in the Wild (CVinW), and is publicly released at at https://computer-vision-in-the-wild.github.io/ELEVATER/