MetaView: Few-shot Active Object Recognition
This addresses a meaningful challenge in robot sensing by enabling efficient view selection with limited data, though it is incremental as it builds on meta-learning for an existing task.
The paper tackles the problem of few-shot active object recognition, where an agent must learn to select informative viewpoints for object recognition from only a few training samples, and demonstrates that their MetaView method outperforms baselines by large margins in experiments.
In robot sensing scenarios, instead of passively utilizing human captured views, an agent should be able to actively choose informative viewpoints of a 3D object as discriminative evidence to boost the recognition accuracy. This task is referred to as active object recognition. Recent works on this task rely on a massive amount of training examples to train an optimal view selection policy. But in realistic robot sensing scenarios, the large-scale training data may not exist and whether the intelligent view selection policy can be still learned from few object samples remains unclear. In this paper, we study this new problem which is extremely challenging but very meaningful in robot sensing -- Few-shot Active Object Recognition, i.e., to learn view selection policies from few object samples, which has not been considered and addressed before. We solve the proposed problem by adopting the framework of meta learning and name our method "MetaView". Extensive experiments on both category-level and instance-level classification tasks demonstrate that the proposed method can efficiently resolve issues that are hard for state-of-the-art active object recognition methods to handle, and outperform several baselines by large margins.