Preprint · 2016 · Open access
Relief R-CNN : Utilizing Convolutional Feature Interrelationship for Object Detection
arXiv (Cornell University) · Published 25 January 2016
In brief
“Relief R-CNN : Utilizing Convolutional Feature Interrelationship for Object Detection” is a 2016 preprint by Guiying Li and 5 co-authors, published in arXiv (Cornell University). OpenAlex counts 2 citations as of 15 November 2026. It is open access: the full text is free to read at arXiv (Cornell University).
- Citations
- 2
- Most cited in
- 2016 (1)
- References
- 41
- Authors
- 6
Abstract
The state-of-the-art object detection pipeline needs a set of object location hypotheses followed by a deep CNN classifier. Previous research on this paradigm, usually extracts features from the same image for the two steps separately, which is time consuming and is hard to optimize. This work shows that the high-level patterns of feature values in deep convolutional feature map contain plenty of useful spatial information and proposes a new deep learning approach to object detection, namely Relief R-CNN ($R^2$-CNN). By extracting positions of objects from these high-level patterns, $R^2$-CNN generates region proposals and performs deep classification simultaneously using the same forward CNN features, unifying the formerly separated object detection process. In this way, $R^2$-CNN does not involve additional information extraction process for region proposal generation, considerably reducing the total computation costs. In addition, a recursive fine-tune technique is also developed for refining coarse proposals. Empirical results showed that our $R^2$-CNN had a very high speed and a reasonable detection rate even in the presence of limit on the proposal number, indicating that the region proposals generated by our $R^2$-CNN are in excellent quality.
Cite this paper
Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.
APA 7
Li, G., Liu, J., Jiang, C., Zhang, L., Tang, K., & Liu, Y. (2016). Relief R-CNN : Utilizing Convolutional Feature Interrelationship for Object Detection. arXiv (Cornell University). https://arxiv.org/pdf/1601.06719
MLA 9
Li, Guiying, et al. “Relief R-CNN : Utilizing Convolutional Feature Interrelationship for Object Detection.” arXiv (Cornell University), 2016. https://arxiv.org/pdf/1601.06719.
Chicago
Li, Guiying, Junlong Liu, Chunhui Jiang, Liangpeng Zhang, Ke Tang, and Yufeng Liu. 2016. “Relief R-CNN : Utilizing Convolutional Feature Interrelationship for Object Detection.” arXiv (Cornell University). https://arxiv.org/pdf/1601.06719.
Harvard
Li, G. et al. (2016) ‘Relief R-CNN : Utilizing Convolutional Feature Interrelationship for Object Detection’, arXiv (Cornell University). Available at: https://arxiv.org/pdf/1601.06719.
IEEE
G. Li, J. Liu, C. Jiang, L. Zhang, K. Tang, and Y. Liu, “Relief R-CNN : Utilizing Convolutional Feature Interrelationship for Object Detection,” arXiv (Cornell University), 2016.
BibTeX
@misc{li2016relief,
title = {{Relief R-CNN : Utilizing Convolutional Feature Interrelationship for Object Detection}},
author = {Li, Guiying and Liu, Junlong and Jiang, Chunhui and Zhang, Liangpeng and Tang, Ke and Liu, Yufeng},
howpublished = {arXiv (Cornell University)},
year = {2016},
url = {https://arxiv.org/pdf/1601.06719}
}
Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)
Topic and keywords
Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- object detection
- R-CNN
- region proposal generation
- proposal refinement
- computational efficiency
Work with this paper
Most-cited papers in Computer Science
- ImageNet classification with deep convolutional neural networksAlex Krizhevsky, Ilya Sutskever, Geoffrey E. Hinton · Communications of the ACM · 2017 · 114,726 citations
- Long Short-Term MemorySepp Hochreiter, Jürgen Schmidhuber · Neural Computation · 1997 · 109,097 citations
- A Mathematical Theory of CommunicationClaude E. Shannon · Bell System Technical Journal · 1948 · 91,736 citations
- Deep learningYann LeCun, Yoshua Bengio, Geoffrey E. Hinton · Nature · 2015 · 90,226 citations
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal NetworksShaoqing Ren, Kaiming He, Ross Girshick and 1 more · IEEE Transactions on Pattern Analysis and Machine Intelligence · 2016 · 74,143 citations
Questions about this paper
How do I cite this paper in APA style?
Li, G., Liu, J., Jiang, C., Zhang, L., Tang, K., & Liu, Y. (2016). Relief R-CNN : Utilizing Convolutional Feature Interrelationship for Object Detection. arXiv (Cornell University). https://arxiv.org/pdf/1601.06719
How many times has this paper been cited?
2 times, according to OpenAlex on 15 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.
Is this paper free to read?
Yes. It is open access: read the free full text at arXiv (Cornell University).
Who wrote this paper?
Guiying Li, Junlong Liu, Chunhui Jiang, Liangpeng Zhang, Ke Tang, Yufeng Liu.
Bibliographic data from OpenAlex (CC0), last checked 15 November 2026. ResearchGuru does not host this paper; the links go to the publisher or the repository that does.