Uncovering the Full Potential of Visual Grounding Methods in VQA
This addresses a methodological issue in VQA research, potentially improving the assessment and development of VG methods, but it is incremental as it focuses on correcting evaluation schemes rather than introducing new methods.
The study tackled the problem that Visual Grounding (VG) methods in Visual Question Answering (VQA) are evaluated under flawed assumptions about the availability of relevant visual information in imperfect image representations, showing that these methods can be much more effective when evaluation conditions are corrected.
Visual Grounding (VG) methods in Visual Question Answering (VQA) attempt to improve VQA performance by strengthening a model's reliance on question-relevant visual information. The presence of such relevant information in the visual input is typically assumed in training and testing. This assumption, however, is inherently flawed when dealing with imperfect image representations common in large-scale VQA, where the information carried by visual features frequently deviates from expected ground-truth contents. As a result, training and testing of VG-methods is performed with largely inaccurate data, which obstructs proper assessment of their potential benefits. In this study, we demonstrate that current evaluation schemes for VG-methods are problematic due to the flawed assumption of availability of relevant visual information. Our experiments show that these methods can be much more effective when evaluation conditions are corrected. Code is provided on GitHub.