The DOI matches its Crossref record; 48 of the 48 references it cites were found in a registry or an index; 3 of the 20 sources its claims cite were checked for support.
Object Detection Networks on Convolutional Feature Maps
Open accessAbstract
Most object detectors contain two important components: a feature extractor and an object classifier. The feature extractor has rapidly evolved with significant research efforts leading to better deep convolutional architectures. The object classifier, however, has not received much attention and many recent systems (like SPPnet and Fast/Faster R-CNN) use simple multi-layer perceptrons. This paper demonstrates that carefully designing deep networks for object classification is just as important. We experiment with region-wise classifier networks that use shared, region-independent convolutional features. We call them "Networks on Convolutional feature maps" (NoCs). We discover that aside from deep feature maps, a deep and convolutional per-region classifier is of particular importance for object detection, whereas latest superior image classification models (such as ResNets and GoogLeNets) do not directly lead to good detection accuracy without using such a per-region classifier. We show by experiments that despite the effective ResNets and Faster R-CNN systems, the design of NoCs is an essential element for the 1st-place winning entries in ImageNet and MS COCO challenges 2015.
Overview
CoChat Summary
TL;DR
The paper studies region-wise classifier networks, called Networks on Convolutional feature maps (NoCs), built on shared convolutional features for object detection. It finds that deep, convolutional per-region classifiers substantially improve detection and are essential for Faster R-CNN with ResNets to achieve strong MS COCO results.
Key findings
- On PASCAL VOC 2007 with ZF features and VOC07 trainval, replacing SVM-on-RoI (45.8% mAP) with deeper MLP NoCs improves mAP: 2fc 49.0, 3fc 53.1, 4fc 53.6; the 4-fc NoC is 7.8% higher than the SVM.
- Adding convolutional layers to NoCs helps with more training data: on VOC07+12, 3fc NoC reaches 56.5, 1conv3fc 58.5, 2conv3fc 58.9, and 3conv3fc 58.8; maxout scale selection raises 2conv3fc from 58.9 to 60.7 (up to 1.8% mAP). Fine-tuning gives 62.9 vs. 60.7 for a randomly initialized NoC and 47.7 for SVM on RoI (ZF, VOC07+12).
- Deep classifiers complement deep features: VGG16 3fc is 64.6, VGG19 3fc 65.1 (+0.5), VGG16 2conv3fc 66.1 (+1.5), and VGG16 maxout 2conv3fc 68.8; with bounding box regression, NoC reaches 71.6 mAP on VOC07 and 68.8 on VOC2012. Error analysis shows stronger NoCs reduce localization error (22.6 vs. 28.1 with ZF; 20.1 vs. 24.8 with VGG16).
- On MS COCO val with Faster R-CNN, naive ResNet-101 has 16.9 AP (stride 32) or 21.3 AP (stride 16); a 3fc MLP NoC raises these to 21.2 and 26.3, while a convolutional NoC reaches 27.2 AP, with AP@0.75 improving from 25.9 to 27.6 and AP@0.5 nearly unchanged from 48.1 to 48.4. This system is described as the foundation of the ResNet paper’s 1st-place MS COCO and ImageNet 2015 entries.
Methods
- Ablation experiments use PASCAL VOC 2007 (20 categories, mAP on 5k test images), training on VOC07 trainval (5k) or VOC07+12 trainval (16k). Shared convolutional features from ImageNet-pretrained ZF and VGG-16/19 are frozen; ~2,000 Selective Search proposals are RoI-pooled to 6×6 (ZF) or 7×7 (VGG); NoCs are trained by backpropagation/SGD, then second-to-last fc features train a linear SVM; no bounding box regression is used in these ablations.
- NoC families include MLPs (2–4 fc layers), ConvNets (1–3 additional 3×3 conv layers plus 3 fc layers), and maxout NoCs for scale selection from two adjacent feature-pyramid scales; fine-tuning initializes fc layers from pretrained weights and conv layers as identity mappings.
- Faster R-CNN + ResNet experiments use MS COCO (80 categories, 80k train, 40k val), reporting COCO-style AP, AP@0.5, and AP@0.75; they compare VGG-16, GoogleNet BN-Inception, and ResNet-101 with end-to-end fine-tuning, RPN proposals, and a softmax classifier plus jointly learned bounding box regressor.
Limitations
- Stated: the ablation experiments are mainly based on SPPnet with frozen shared feature maps, precomputed Selective Search proposals, and a post-hoc SVM; the authors say observations generally hold when these restricted conditions are relaxed or removed, as shown with Faster R-CNN. They also state VOC07 trainval is too small to train deeper models, causing overfitting, and that no bounding box regression is used in the ablation tables.
- Stated: differences among maxout variants are marginal, and the benefit of using a ConvNet as NoC diminishes for Faster R-CNN + ResNet on PASCAL VOC (both MLP and ConvNet NoCs reach 76.4 vs. 71.9 for the 1fc baseline).
- Evident from the text: evaluation is limited to PASCAL VOC and MS COCO with specific backbones (ZF, VGG, GoogleNet, ResNet), so broader datasets or architectures are not tested.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
2 Valid · 0 Not valid · 13 Not checked · 34 Own results and interpretations
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“However, for 'fully convolutional' models such as GoogleNets [18] and ResNets [17], there are no hidden fc layers for building regionwise classifiers.”
What the sources say
- Supportedcites [18] [18] C. Szegedy, W. Liu, Y. Jia, P. Sermanet, S. Reed, D. Anguelov, D. Erhan, and A. Rabinovich, 'Going deeper with convolutions,' Tech. Rep., 2014. [Online]. Available: http://arxiv.org/pdf/1409.4842v1 - source
The GoogLeNet section (p. 6) says the main classifier replaces fully connected layers with average pooling, retaining an output linear layer. The Detection section (p. 10) says GoogLeNet uses the Inception model as a region classifier. The paper does describe a hidden fully connected layer in an auxiliary training classifier, but says that classifier is discarded at inference time; the separate ResNet citation is outside this verdict. (LLM, conf=0.85)
- Could not verifycites [17] [17] K. He, X. Zhang, S. Ren, and J. Sun, 'Deep residual learning for image recognition,' in arXiv prepring arXiv:1506.01497 , 2015. - sourceExperimental — accuracy not yet measured
Section 3.1 describes a fully convolutional region proposal network and its classification layer, but does not say whether GoogLeNet or ResNet has hidden fully connected layers for regionwise classifiers. Searches for those architectures and hidden fully connected layers found no passage in our copy that addresses the claim. (LLM, conf=0.72)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“The above system (27.2% AP and 48.4% AP@0.5) is the foundation of the detection system in the ResNet paper [17].”
What the sources say
- Supportedcites [17] [17] K. He, X. Zhang, S. Ren, and J. Sun, 'Deep residual learning for image recognition,' in arXiv prepring arXiv:1506.01497 , 2015. - source
The paper’s Additional information section says that replacing VGG-16 with ResNet-101 in Faster R-CNN yields 48.4%/27.2% on the COCO validation set. It also describes the ResNet authors’ further improvements as orthogonal to Faster R-CNN, supporting the characterization of Faster R-CNN as the foundation of their detection system. (LLM, conf=0.90)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Our best NoC using VGG-16 has 68.8% mAP (Table 5).”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“Fig. 3 shows that VGG-16 in general has lower recognition error than the ZF net, when using the same classifiers ( e.g ., 1.6%+1.3%+7.4% vs . 3.2%+2.2%+7.4%).”
What the sources say
- Not checked yetcites [30] [30] D. Hoiem, Y. Chodpathumwan, and Q. Dai, 'Diagnosing error in object detectors,' in ECCV , 2012. -
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“On the other hand, when the depth of region-aware classifier is increased (but still using the VGG-16 features), the mAP is increased by 1.5% to 66.1%.”
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
22 more claims
Create a free account to see all 49 claims and to verify the 12 references still open.
- It may state two separate claims as one. 18
- Probably not a substantial claim. 8
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Experimental Results
“In all cases, we set the output resolution of RoI pooling as 7 × 7. This is followed by a 81-d classifier (equivalent to a 1fc NoC).”
§ Experimental Results
“We argue that this is because in the case of na¨ ıve Faster R-CNN, VGG-16 has a 3fc NoC but GoogleNet and ResNet has a 1fc NoC (Table 8).”
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474
Self-citationVery Deep Convolutional Networks for Large-Scale Image Recognition
Simonyan, K., & Zisserman, A. (2014) · arXiv · doi:10.48550/arxiv.1409.1556 · Cited by 114,148
ImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155
Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Ren, S., He, K., Girshick, R., & Sun, J. (2017) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2016.2577031 · Cited by 73,472
Self-citationFaster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Ren, S., He, K., Girshick, R., & Sun, J. (2015) · arXiv · doi:10.48550/arxiv.1506.01497 · Cited by 70,812
Self-citationImageNet: A large-scale hierarchical image database
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009) · 2009 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2009.5206848 · Cited by 65,333
Distinctive Image Features from Scale-Invariant Keypoints
Lowe, D. G. (2004) · International Journal of Computer Vision · doi:10.1023/b:visi.0000029664.99615.94 · Cited by 55,589
Going deeper with convolutions
Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., & Rabinovich, A. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298594 · Cited by 48,562
Microsoft COCO: Common Objects in Context
Lin, T.-Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollár, P., & Zitnick, C. L. (2014) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-10602-1_48 · Cited by 45,460
Fully convolutional networks for semantic segmentation
Long, J., Shelhamer, E., & Darrell, T. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298965 · Cited by 38,834
Research works printed but not linked: 2
- goodfellow, 2013, Maxout networks, ICML (3), 28, 1319
- sermanet, 0, OverFeat: Integrated recognition, localization and detection using convolutional networks, International Conference on Learning Representations (ICLR 2014)
Cited by
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474
Self-citationFaster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Ren, S., He, K., Girshick, R., & Sun, J. (2015) · arXiv · doi:10.48550/arxiv.1506.01497 · Cited by 70,812
Self-citationFaster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Ren, S., He, K., Girshick, R., & Sun, J. (2017) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2016.2577031 · Cited by 73,472
Self-citationYou Only Look Once: Unified, Real-Time Object Detection
Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.91 · Cited by 53,803
Self-citationFeature Pyramid Networks for Object Detection
Lin, T.-Y., Dollar, P., Girshick, R., He, K., Hariharan, B., & Belongie, S. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.106 · Cited by 30,939
Self-citation
Metadata
- Authors
- Ren, S., He, K., Girshick, R., Zhang, X., & Sun, J.
- Affiliations
- University of Science and Technology of China (CN)
- Meta (United States) (US)
- Xi'an Jiaotong University (CN)
- Venue
- IEEE Transactions on Pattern Analysis and Machine Intelligence
- Published
- 2017-07-01
- Volume
- 39
- Issue
- 7
- Pages
- 1476-1481
- ISSN
- 0162-8828 (ISSN-L)
- 2160-9292
- OpenAlex
- W2117287331
- Semantic Scholar
- f075f89b4f4026748cbf2fb9f989a9934c42ee8f
- Cited by
- 454
- References
- 48
- Citations by year
- 2026: 6 · 2025: 19 · 2024: 28 · 2023: 46 · 2022: 41 · 2021: 64 · 2020: 59 · 2019: 76 · 2018: 48 · 2017: 41
- FWCI
- 42.31
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- IEEE Transactions on Pattern Analysis and Machine Intelligence (publishedVersion)
- PubMed (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Image and Video Retrieval Techniques
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Visual Attention and Saliency Detection
- Keywords
- object detection, object classification, Faster R-CNN, ResNet, GoogLeNet, feature extractor, MS COCO, ImageNet
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-09-27, under version 3 of the claim definition (prompt extract-v3).
A second model labelled each claim on 2026-09-27 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Object Detection Networks on Convolutional Feature Maps. https://app.cochat.ai/papers/10.1109/tpami.2016.2601099
- The paper
- Ren, S., He, K., Girshick, R., Zhang, X., & Sun, J. (2017). Object Detection Networks on Convolutional Feature Maps. IEEE Transactions on Pattern Analysis and Machine Intelligence. https://doi.org/10.1109/tpami.2016.2601099
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.