Recurrent Feedback Improves Recognition of Partially Occluded Objects
This work addresses the challenge of robust object recognition in computer vision, particularly for occluded objects, with incremental improvements in model design.
The study tackled the problem of object recognition under occlusion by investigating whether artificial neural networks with recurrent connectivity improve performance, finding that recurrent models achieved significantly higher classification accuracy than feedforward models of matched complexity and could revise initial guesses for challenging stimuli.
Recurrent connectivity in the visual cortex is believed to aid object recognition for challenging conditions such as occlusion. Here we investigate if and how artificial neural networks also benefit from recurrence. We compare architectures composed of bottom-up, lateral and top-down connections and evaluate their performance using two novel stereoscopic occluded object datasets. We find that classification accuracy is significantly higher for recurrent models when compared to feedforward models of matched parametric complexity. Additionally we show that for challenging stimuli, the recurrent feedback is able to correctly revise the initial feedforward guess.