Real-world Object Recognition with Off-the-shelf Deep Conv Nets: How Many Objects can iCub Learn?
This work addresses the bottleneck of visual perception for robotics, though it is incremental as it applies existing computer vision methods to a new robotic dataset.
The study investigated the generalization of deep convolutional networks to robotic object recognition using the iCub humanoid robot, finding that off-the-shelf models achieved competitive performance on the new iCubWorld28 dataset but revealed limitations in real-world robotic settings.
The ability to visually recognize objects is a fundamental skill for robotics systems. Indeed, a large variety of tasks involving manipulation, navigation or interaction with other agents, deeply depends on the accurate understanding of the visual scene. Yet, at the time being, robots are lacking good visual perceptual systems, which often become the main bottleneck preventing the use of autonomous agents for real-world applications. Lately in computer vision, systems that learn suitable visual representations and based on multi-layer deep convolutional networks are showing remarkable performance in tasks such as large-scale visual recognition and image retrieval. To this regard, it is natural to ask whether such remarkable performance would generalize also to the robotic setting. In this paper we investigate such possibility, while taking further steps in developing a computational vision system to be embedded on a robotic platform, the iCub humanoid robot. In particular, we release a new dataset ({\sc iCubWorld28}) that we use as a benchmark to address the question: {\it how many objects can iCub recognize?} Our study is developed in a learning framework which reflects the typical visual experience of a humanoid robot like the iCub. Experiments shed interesting insights on the strength and weaknesses of current computer vision approaches applied in real robotic settings.