CLIPping the Limits: Finding the Sweet Spot for Relevant Images in Automated Driving Systems Perception Testing
This work addresses the challenge of automating image retrieval for perception testing in automated driving systems, representing an incremental improvement in dataset management.
The paper tackles the problem of efficiently finding specific images in large datasets for testing automated driving systems by automating the threshold definition in sorted results from natural language prompts, aiming to return only relevant images while preventing false positives and false negatives.
Perception systems, especially cameras, are the eyes of automated driving systems. Ensuring that they function reliably and robustly is therefore an important building block in the automation of vehicles. There are various approaches to test the perception of automated driving systems. Ultimately, however, it always comes down to the investigation of the behavior of perception systems under specific input data. Camera images are a crucial part of the input data. Image data sets are therefore collected for the testing of automated driving systems, but it is non-trivial to find specific images in these data sets. Thanks to recent developments in neural networks, there are now methods for sorting the images in a data set according to their similarity to a prompt in natural language. In order to further automate the provision of search results, we make a contribution by automating the threshold definition in these sorted results and returning only the images relevant to the prompt as a result. Our focus is on preventing false positives and false negatives equally. It is also important that our method is robust and in the case that our assumptions are not fulfilled, we provide a fallback solution.