Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Object Detection Based on Distributed Convolutional Neural Networks

Liang Sun

Abstract

Based on the Distributed Convolutional Neural Network(DisCNN), a straightforward object detection method is proposed. The modules of the output vector of a DisCNN with respect to a specific positive class are positively monotonic with the presence probabilities of the positive features. So, by identifying all high-scoring patches across all possible scales, the positive object can be detected by overlapping them to form a bounding box. The essential idea is that the object is detected by detecting its features on multiple scales, ranging from specific sub-features to abstract features composed of these sub-features. Training DisCNN requires only object-centered image data with positive and negative class labels. The detection process for multiple positive classes can be conducted in parallel to significantly accelerate it, and also faster for single-object detection because of its lightweight model architecture.

Object Detection Based on Distributed Convolutional Neural Networks

Abstract

Based on the Distributed Convolutional Neural Network(DisCNN), a straightforward object detection method is proposed. The modules of the output vector of a DisCNN with respect to a specific positive class are positively monotonic with the presence probabilities of the positive features. So, by identifying all high-scoring patches across all possible scales, the positive object can be detected by overlapping them to form a bounding box. The essential idea is that the object is detected by detecting its features on multiple scales, ranging from specific sub-features to abstract features composed of these sub-features. Training DisCNN requires only object-centered image data with positive and negative class labels. The detection process for multiple positive classes can be conducted in parallel to significantly accelerate it, and also faster for single-object detection because of its lightweight model architecture.

Paper Structure

This paper contains 10 sections, 6 theorems, 4 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. Model and training
  3. DisCNN architecture
  4. Training with N2O loss
  5. Object detection
  6. Object detection algorithm
  7. Module is a metric for the presence probability of positive features
  8. One-car scenario: object detection via feature detection
  9. Multi-car scenario
  10. Conclusion

Key Result

Lemma 1

The H2O loss requires that the positive class have no features in common with the negative class in the training data.

Figures (4)

  • Figure 1: Patches sorted by modules
  • Figure 2: One-car detection
  • Figure 3: Car sub-feature detection
  • Figure 4: multi-car detection

Theorems & Definitions (8)

  • Lemma 1
  • Lemma 2
  • Lemma 3
  • Remark 1
  • Remark 2
  • Proposition 1
  • Proposition 2
  • Proposition 3