Untangling Local and Global Deformations in Deep Convolutional Networks for Image Classification and Sliding Window Detection
This work addresses the challenge of deformation invariance in image classification and object detection for computer vision applications, offering incremental improvements over existing methods.
The paper tackled the problem of handling deformations in deep convolutional networks by introducing epitomic convolution and a multiple instance learning approach for global deformations, resulting in faster convergence, better generalization, and substantial classification accuracy improvements on ImageNet and Pascal VOC 2007 benchmarks.
Deep Convolutional Neural Networks (DCNNs) commonly use generic `max-pooling' (MP) layers to extract deformation-invariant features, but we argue in favor of a more refined treatment. First, we introduce epitomic convolution as a building block alternative to the common convolution-MP cascade of DCNNs; while having identical complexity to MP, Epitomic Convolution allows for parameter sharing across different filters, resulting in faster convergence and better generalization. Second, we introduce a Multiple Instance Learning approach to explicitly accommodate global translation and scaling when training a DCNN exclusively with class labels. For this we rely on a `patchwork' data structure that efficiently lays out all image scales and positions as candidates to a DCNN. Factoring global and local deformations allows a DCNN to `focus its resources' on the treatment of non-rigid deformations and yields a substantial classification accuracy improvement. Third, further pursuing this idea, we develop an efficient DCNN sliding window object detector that employs explicit search over position, scale, and aspect ratio. We provide competitive image classification and localization results on the ImageNet dataset and object detection results on the Pascal VOC 2007 benchmark.