Dipartimento di Ingegneria informatica, automatica e gestionale

The ability to classify objects is fundamental for robots. Besides knowledge about their visual appearance, captured by the RGB channel, robots heavily need also depth information to make sense of the world. While the use of deep networks on RGB robot images has benefited from the plethora of results obtained on databases like ImageNet, using convnets on depth images requires mapping them into three-dimensional channels. This transfer learning proceduremakes them processable by pretrained deep architectures. Current mappings are based on heuristic assumptions over preprocessing steps and on what depth properties should be most preserved, resulting often in cumbersome data visualizations, and in suboptimal performance in terms of generality and recognition results. Here, we take an alternative route and we attempt instead to learn an optimal colorization mapping for any given pretrained architecture, using as training data a reference RGB-D database. We propose a deep network architecture, exploiting the residual paradigm, that learns how to map depth data to three channel images. A qualitative analysis of the images obtained with this approach clearly indicates that learning the optimal mapping preserves the richness of depth information better than current hand-crafted approaches. Experiments on the Washington, JHUIT-50 and BigBIRD public benchmark databases, using CaffeNet, VGG-16, GoogleNet, and ResNet50 clearly showcase the power of our approach, with gains in performance of up to 16% compared to state of the art competitors on the depth channel only, leading to top performances when dealing with RGB-D data.