Learning to count with deep object features

Abstract

Learning to count is a learning strategy that has been recently proposed in the literature for dealing with problems where estimating the number of object instances in a scene is the final objective. In this framework, the task of learning to detect and localize individual object instances is seen as a harder task that can be evaded by casting the problem as that of computing a regression value from hand-crafted image features. In this paper we explore the features that are learned when training a counting convolutional neural network in order to understand their underlying representation. To this end we define a counting problem for MNIST data and show that the internal representation of the network is able to classify digits in spite of the fact that no direct supervision was provided for them during training. We also present preliminary results about a deep network that is able to count the number of pedestrians in a scene.

This paper has been accepted at Deep Vision Workshop at CVPR 2015

References in corpus (3)

• Caffe: Convolutional Architecture for Fast Feature Embedding
• Object Detectors Emerge in Deep Scene CNNs
• Hypercolumns for Object Segmentation and Fine-grained Localization

Cited by in corpus (9)

• Classification vs. Regression in Supervised Learning for Single Channel Speaker Count Estimation
• Improving Object Counting with Heatmap Regulation
• Record Counting in Historical Handwritten Documents with Convolutional Neural Networks
• Textually Enriched Neural Module Networks for Visual Question Answering
• Learning to Point and Count
• Systematically designing better instance counting models on cell images with Neural Arithmetic Logic Units
• Reliable counting of weakly labeled concepts by a single spiking neuron model
• Pay Attention to Those Sets! Learning Quantification from Images
• Be Precise or Fuzzy: Learning the Meaning of Cardinals and Quantifiers from Vision