Ambiguous Images With Human Judgments for Robust Visual Event Classification
This addresses the issue of robust visual event classification for AI systems by introducing a dataset to evaluate and improve model performance on noisy, real-world data, though it is incremental as it builds on existing vision benchmarks.
The authors tackled the problem of vision models performing poorly on ambiguous images by creating SQUID-E, a dataset of noisy images with human uncertainty annotations. They found that existing models fail to provide meaningful outputs for such data, motivating further research on ambiguous datasets.
Contemporary vision benchmarks predominantly consider tasks on which humans can achieve near-perfect performance. However, humans are frequently presented with visual data that they cannot classify with 100% certainty, and models trained on standard vision benchmarks achieve low performance when evaluated on this data. To address this issue, we introduce a procedure for creating datasets of ambiguous images and use it to produce SQUID-E ("Squidy"), a collection of noisy images extracted from videos. All images are annotated with ground truth values and a test set is annotated with human uncertainty judgments. We use this dataset to characterize human uncertainty in vision tasks and evaluate existing visual event classification models. Experimental results suggest that existing vision models are not sufficiently equipped to provide meaningful outputs for ambiguous images and that datasets of this nature can be used to assess and improve such models through model training and direct evaluation of model calibration. These findings motivate large-scale ambiguous dataset creation and further research focusing on noisy visual data.