Multi-Scale Dense Networks for Resource Efficient Image Classification

In this paper we investigate image classification with computational resource limits at test time. Two such settings are: 1. anytime classification, where the network's prediction for a test example is progressively updated, facilitating the output of a prediction at any time; and 2. budgeted batch classification, where a fixed amount of computation is available to classify a set of examples that can be spent unevenly across "easier" and "harder" inputs. In contrast to most prior work, such as the popular Viola and Jones algorithm, our approach is based on convolutional neural networks. We train multiple classifiers with varying resource demands, which we adaptively apply during test time. To maximally re-use computation between the classifiers, we incorporate them as early-exits into a single deep convolutional neural network and inter-connect them with dense connectivity. To facilitate high quality classification early on, we use a two-dimensional multi-scale network architecture that maintains coarse and fine level features all-throughout the network. Experiments on three image-classification tasks demonstrate that our framework substantially improves the existing state-of-the-art in both settings.

PDF Abstract ICLR 2018 PDF ICLR 2018 Abstract
Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Handwritten Mathmatical Expression Recognition CROHME 2016 DenseWAP-MSA ExpRate 50.1 # 12
Handwritten Mathmatical Expression Recognition CROHME 2016 DenseWAP ExpRate 47.5 # 14
Handwritten Mathmatical Expression Recognition CROHME 2019 DenseWAP-MSA ExpRate 47.7 # 13
Handwritten Mathmatical Expression Recognition HME100K DenseWAP ExpRate 61.85 # 12

Methods


No methods listed for this paper. Add relevant methods here