ACP++: Action Co-occurrence Priors for Human-Object Interaction Detection
This addresses the challenge of detecting rare human-object interactions for computer vision applications, representing an incremental improvement by leveraging prior knowledge to enhance training.
The paper tackles the problem of low classification accuracy for rare human-object interaction (HOI) classes due to long-tailed training sets by modeling natural correlations and anti-correlations as action co-occurrence priors, resulting in consistent performance improvements over state-of-the-art methods on HICO-Det and V-COCO benchmarks.
A common problem in the task of human-object interaction (HOI) detection is that numerous HOI classes have only a small number of labeled examples, resulting in training sets with a long-tailed distribution. The lack of positive labels can lead to low classification accuracy for these classes. Towards addressing this issue, we observe that there exist natural correlations and anti-correlations among human-object interactions. In this paper, we model the correlations as action co-occurrence matrices and present techniques to learn these priors and leverage them for more effective training, especially on rare classes. The efficacy of our approach is demonstrated experimentally, where the performance of our approach consistently improves over the state-of-the-art methods on both of the two leading HOI detection benchmark datasets, HICO-Det and V-COCO.