Faster R-CNN: Towards Real-Time Object Detection

with Region Proposal Networks

Shaoqing Ren∗ Kaiming He Ross Girshick Jian Sun

Microsoft Research
{v-shren, kahe, rbg, jiansun}@[Link]
arXiv:1506.01497v1 [[Link]] 4 Jun 2015

Abstract
State-of-the-art object detection networks depend on region proposal algorithms
to hypothesize object locations. Advances like SPPnet [7] and Fast R-CNN [5]
have reduced the running time of these detection networks, exposing region pro-
posal computation as a bottleneck. In this work, we introduce a Region Proposal
Network (RPN) that shares full-image convolutional features with the detection
network, thus enabling nearly cost-free region proposals. An RPN is a fully-
convolutional network that simultaneously predicts object bounds and objectness
scores at each position. RPNs are trained end-to-end to generate high-quality
region proposals, which are used by Fast R-CNN for detection. With a simple
alternating optimization, RPN and Fast R-CNN can be trained to share convolu-
tional features. For the very deep VGG-16 model [18], our detection system has a
frame rate of 5fps (including all steps) on a GPU, while achieving state-of-the-art
object detection accuracy on PASCAL VOC 2007 (73.2% mAP) and 2012 (70.4%
mAP) using 300 proposals per image. The code will be released.

1 Introduction

Recent advances in object detection are driven by the success of region proposal methods (e.g., [21])
and region-based convolutional neural networks (R-CNNs) [6]. Although region-based CNNs were
computationally expensive as originally developed in [6], their cost has been drastically reduced
thanks to sharing convolutions across proposals [7, 5]. The latest incarnation, Fast R-CNN [5],
achieves near real-time rates using very deep networks [18], when ignoring the time spent on region
proposals. Now, proposals are the computational bottleneck in state-of-the-art detection systems.
Region proposal methods typically rely on inexpensive features and economical inference schemes.
Selective Search (SS) [21], one of the most popular methods, greedily merges superpixels based
on engineered low-level features. Yet when compared to efficient detection networks [5], Selective
Search is an order of magnitude slower, at 2s per image in a CPU implementation. EdgeBoxes
[23] currently provides the best tradeoff between proposal quality and speed, at 0.2s per image.
Nevertheless, the region proposal step still consumes as much running time as the detection network.
One may note that fast region-based CNNs take advantage of GPUs, while the region proposal meth-
ods used in research are implemented on the CPU, making such runtime comparisons inequitable.
An obvious way to accelerate proposal computation is to re-implement it for the GPU. This may be
an effective engineering solution, but re-implementation ignores the down-stream detection network
and therefore misses important opportunities for sharing computation.
In this paper, we show that an algorithmic change—computing proposals with a deep net—leads
to an elegant and effective solution, where proposal computation is nearly cost-free given the de-
tection network’s computation. To this end, we introduce novel Region Proposal Networks (RPNs)
∗
Shaoqing Ren is with the University of Science and Technology of China. This work was done when he
was an intern at Microsoft Research.

1
that share convolutional layers with state-of-the-art object detection networks [7, 5]. By sharing
convolutions at test-time, the marginal cost for computing proposals is small (e.g., 10ms per image).
Our observation is that the convolutional (conv) feature maps used by region-based detectors, like
Fast R-CNN, can also be used for generating region proposals. On top of these conv features,
we construct RPNs by adding two additional convolutional layers: one that encodes each conv map
position into a short (e.g., 256-d) feature vector and a second that, at each conv map position, outputs
an objectness score and regressed bounds for k region proposals relative to various scales and aspect
ratios at that location (k = 9 is a typical value).
Our RPNs are thus a kind of fully-convolutional network (FCN) [13] and they can be trained end-to-
end specifically for the task for generating detection proposals. To unify RPNs with Fast R-CNN [5]
object detection networks, we propose a simple training scheme that alternates between fine-tuning
for the region proposal task and then fine-tuning for object detection, while keeping the proposals
fixed. This scheme converges quickly and produces a unified network with convolutional features
that are shared between both tasks.
We evaluate our method on the PASCAL VOC detection benchmarks [4], where RPNs with Fast
R-CNNs produce detection accuracy better than the strong baseline of Selective Search with Fast
R-CNNs. Meanwhile, our method waives nearly all computational burdens of SS at test-time—the
effective running time for proposals is just 10 milliseconds. Using the expensive very deep models
of [18], our detection method still has a frame rate of 5fps (including all steps) on a GPU, and thus
is a practical object detection system in terms of both speed and accuracy (73.2% mAP on PASCAL
VOC 2007 and 70.4% mAP on 2012). The code will be released.

2 Related Work

Several recent papers have proposed ways of using deep networks for locating class-specific or
class-agnostic bounding boxes [20, 17, 3, 19]. In the OverFeat method [17], a 4-d output fc layer
(for each or all classes) is trained to predict the box coordinates for the localization task (which
assumes a single object). The fc layer is then turned into a conv layer for detecting multiple class-
specific objects. The MultiBox methods [3, 19] generate region proposals from a network whose
last fc layer simultaneously predicts multiple (e.g., 800) boxes, which are used for R-CNN [6] object
detection. Their proposal network is applied on a single image or multiple large image crops (e.g.,
224×224) [19]. We discuss OverFeat and MultiBox in more depth later in context with our method.
Shared computation of convolutions [17, 7, 2, 5] has been attracting increasing attention for efficient,
yet accurate, visual recognition. The OverFeat paper [17] computes conv features from an image
pyramid for classification, localization, and detection. Adaptively-sized pooling [7] on shared conv
feature maps is proposed for efficient region-based object detection [7, 15] and semantic segmenta-
tion [2]. Fast R-CNN [5] enables end-to-end training of adaptive pooling on shared conv features
and shows compelling accuracy and speed.

3 Region Proposal Networks

A Region Proposal Network (RPN) takes an image (of any size) as input and outputs a set of
rectangular object proposals, each with an objectness score.1 We model this process with a fully-
convolutional network [13], which we describe in this section. Because our ultimate goal is to share
computation with a Fast R-CNN object detection network [5], we assume that both nets share a
common set of conv layers. In our experiments, we investigate the Zeiler and Fergus model [22]
(ZF), which has 5 shareable conv layers and the Simonyan and Zisserman model [18] (VGG), which
has 13 shareable conv layers.
To generate region proposals, we slide a small network over the conv feature map output by the last
shared conv layer. This network is fully connected to an n × n spatial window of the input conv
feature map. Each sliding window is mapped to a lower-dimensional vector (256-d for ZF and 512-d
for VGG). This vector is fed into two sibling fully-connected layers—a box-regression layer (reg)
1
“Region” is a generic term and in this paper we only consider rectangular regions, as is common for many
methods (e.g., [19, 21, 23]). “Objectness” measures membership to a set of object classes vs. background.

2
 person : 0.992
2k scores 4k coordinates k anchor boxes dog : 0.994
horse : 0.993

cls layer reg layer car : 1.000 cat : 0.982

dog : 0.997 person : 0.979

256-d
intermediate layer

bus : 0.996
boat : 0.970

person : 0.983
person : 0.736 person : 0.983
person : 0.925

person : 0.989

sliding window
conv feature map

Figure 1: Left: Region Proposal Network (RPN). Right: Example detections using RPN proposals
on PASCAL VOC 2007 test. Our method detects objects in a wide range of scales and aspect ratios.

and a box-classification layer (cls). We use n = 3 in this paper, noting that the effective receptive
field on the input image is large (171 and 228 pixels for ZF and VGG, respectively). This mini-
network is illustrated at a single position in Fig. 1 (left). Note that because the mini-network operates
in a sliding-window fashion, the fully-connected layers are shared across all spatial locations. This
architecture is naturally implemented with an n × n conv layer followed by two sibling 1 × 1 conv
layers (for reg and cls, respectively). ReLUs [14] are applied to the output of the n × n conv layer.

Translation-Invariant Anchors
At each sliding-window location, we simultaneously predict k region proposals, so the reg layer
has 4k outputs encoding the coordinates of k boxes. The cls layer outputs 2k scores that estimate
probability of object / not-object for each proposal.2 The k proposals are parameterized relative to
k reference boxes, called anchors. Each anchor is centered at the sliding window in question, and is
associated with a scale and aspect ratio. We use 3 scales and 3 aspect ratios, yielding k = 9 anchors
at each sliding position. For a conv feature map of a size W × H, there are W Hk anchors in total.
An important property of our approach is that it is translation invariant, both in terms of the anchors
and the functions that compute proposals relative to the anchors.
As a comparison, the MultiBox method [19] uses k-means to generate 800 anchors, which are not
translation invariant. If one translates an object in an image, the proposal should translate and the
same function should be able to predict the proposal in either location. Moreover, because the
MultiBox anchors are not translation invariant, it requires a (4+1)×800-dimensional output layer,
whereas our method requires a (4+2)×9-dimensional output layer. Our proposal layers have an order
of magnitude fewer parameters (27 million for MultiBox using GoogLeNet [19] vs. 2.4 million for
RPN using VGG-16), and thus have less risk of overfitting on small datasets, like PASCAL VOC.

A Loss Function for Learning Region Proposals

For training RPNs, we assign a binary class label (of being an object or not) to each anchor. We
assign a positive label to two kinds of anchors: (i) the anchor that has the highest Intersection-over-
Union (IoU) overlap with a ground-truth box, or (ii) an anchor that has an IoU overlap higher than
0.7 with any ground-truth box.3 Note that a single ground-truth box may assign positive labels to
multiple anchors. We assign a negative label to a non-positive anchor if its IoU ratio is lower than
0.3 for all ground-truth boxes. There are anchors that are neither positive nor negative, and they are
ignored and will not contribute to the training objective.
With these definitions, we minimize an objective function following the multi-task loss in Fast R-
CNN [5]. For an anchor box i, its loss function is defined as:
L(pi , ti ) = Lcls (pi , p∗i ) + λp∗i Lreg (ti , t∗i ). (1)
2
For simplicity we implement the cls layer as a two-class softmax layer. Alternatively, one may use logistic
regression to produce k scores.
3
We intend to simplify this to case (ii), only. Current experiments, however, consider an anchor as positive
if it satisfies condition (i) or condition (ii).

3
Here pi is the predicted probability of the anchor i being an object. p∗i is 1 if the anchor is labeled
positive, and is 0 if the anchor is negative. ti = {tx , ty , tw , th }i represents the 4 parameterized
coordinates of the predicted bounding box, and t∗i = {t∗x , t∗y , t∗w , t∗h }i represents the ground-truth
box associated with a positive anchor. The classification loss Lcls is the softmax loss of two classes
(object vs. not object). For the regression loss, we use Lreg (ti , t∗i ) = R(ti − t∗i ) where R is the
robust loss function (smooth-L1 ) defined in [5]. The term p∗i Lreg means the regression loss is
activated only for positive anchors (p∗i = 1) and is disabled otherwise (p∗i = 0). The loss-balancing
parameter λ is set to 10, which means that we bias towards better box locations. The outputs of the
cls and reg layers consist of {pi } and {ti } respectively.
We adopt the parameterizations of the 4 coordinates following [6]:
tx = (x − xa )/wa , ty = (y − ya )/ha , tw = log(w/wa ), th = log(h/ha )
t∗x = (x∗ − xa )/wa , t∗y = (y ∗ − ya )/ha , t∗w = log(w∗ /wa ), t∗h = log(h∗ /ha ),
where x, xa , and x∗ are for the predicted box, anchor box, and ground-truth box respectively. This
can be thought of as bounding-box regression from an anchor box to a nearby ground-truth box.
Nevertheless, our method achieves bounding-box regression by a different manner from previous
feature-map-based methods [7, 5]. In [7, 5], bounding-box regression is performed on features
pooled from arbitrarily sized regions, and the regression weights are shared by all region sizes. In
our formulation, the features used for regression are of the same spatial size (n × n) on the feature
maps. To account for varying sizes, a set of k bounding-box regressors are learned. Each regressor
is responsible for one scale and one aspect ratio, and the k regressors do not share weights. As such,
it is still possible to predict boxes of various sizes even though the features are of a fixed size/scale.

Optimization
The RPN, which is naturally implemented as a fully-convolutional network [13], can be trained
end-to-end by back-propagation and stochastic gradient descent (SGD) [12]. We follow the “image-
centric” sampling strategy from [5] to train this network. Each mini-batch arises from a single image
that contains many positive and negative examples. It is possible to optimize for the loss functions of
all anchors, but this will bias towards negative samples as they are dominate. Instead, we randomly
sample 256 anchors in an image to compute the loss function of a mini-batch, where the sampled
positive and negative anchors have a ratio of 1:1.
We randomly initialize all new layers by drawing weights from a zero-mean Gaussian distribution
with standard deviation 0.01. All other layers (i.e., the shared conv layers) are initialized by pre-
training a model for ImageNet classification [16], as is standard practice [6]. We tune all layers of
the ZF net, and conv3 1 and up for the VGG net to conserve memory [5]. We use a learning rate of
0.001 for 60k mini-batches, and 0.0001 for the next 20k mini-batches on the PASCAL dataset. We
also use a momentum of 0.9 and a weight decay of 0.0005 [11]. We implement in Caffe [10].

Sharing Convolutional Features for Region Proposal and Object Detection

Thus far we have described how to train a network for region proposal generation, without con-
sidering the region-based object detection CNN that will utilize these proposals. For the detection
network, we adopt Fast R-CNN [5]4 and now describe an algorithm that learns conv layers that are
shared between the Region Proposal Network and Fast R-CNN.
Both RPN and Fast R-CNN, trained independently, will modify their conv layers in different ways.
We therefore need to develop a technique that allows for sharing conv layers between the two net-
works, rather than learning two separate networks. Note that this is not as easy as simply defining
a single network that includes both RPN and Fast R-CNN, and then optimizing it jointly with back-
propagation. The reason is that Fast R-CNN training depends on fixed object proposals and it is
not clear a priori if learning Fast R-CNN while simultaneously changing the proposal mechanism
will converge. While this joint optimizing is an interesting question for future work, we develop a
pragmatic 4-step training algorithm to learn shared features via alternating optimization.
In the first step, we train the RPN as described above. This network is initialized with an ImageNet-
pre-trained model and fine-tuned end-to-end for the region proposal task. In the second step, we
4
[Link]

4
train a separate detection network by Fast R-CNN using the proposals generated by the step-1 RPN.
This detection network is also initialized by the ImageNet-pre-trained model. At this point the two
networks do not share conv layers. In the third step, we use the detector network to initialize RPN
training, but we fix the shared conv layers and only fine-tune the layers unique to RPN. Now the two
networks share conv layers. Finally, keeping the shared conv layers fixed, we fine-tune the fc layers
of the Fast R-CNN. As such, both networks share the same conv layers and form a unified network.

Implementation Details
We train and test both region proposal and object detection networks on single-scale images [7,
5]. We re-scale the images such that their shorter side is s = 600 pixels [5]. Multi-scale feature
extraction may improve accuracy but does not exhibit a good speed-accuracy trade-off [5].
For anchors, we use 3 scales with box areas of 1282 , 2562 , and 5122 pixels, and 3 aspect ratios of
1:1, 1:2, and 2:1. We note that our algorithm allows use of anchor boxes that are larger than the
underlying receptive field when predicting large proposals. Such predictions are not impossible—
one may still roughly infer the extent of an object if only the middle of the object is visible. With
this design, our solution does not need multi-scale features or multi-scale sliding windows to predict
large regions, saving considerable running time. Fig. 1 (right) shows the capability of our method
for a wide range of scales and aspect ratios. The table below shows the learned average proposal
size for each anchor using the ZF net (numbers for s = 600).
anchor 1282 , 2:1 1282 , 1:1 1282 , 1:2 2562 , 2:1 2562 , 1:1 2562 , 1:2 5122 , 2:1 5122 , 1:1 5122 , 1:2
proposal 188×111 113×114 70×92 416×229 261×284 174×332 768×437 499×501 355×715

The anchor boxes that cross image boundaries need to be handled with care. During training, we
ignore all cross-boundary anchors, and they will not contribute to the loss. For a typical 1000 × 600
image, there will be roughly 20k (≈ 60 × 40 × 9) anchors in total. With the cross-boundary anchors
ignored, there are about 6k anchors per image for training. If the outliers are not ignored in training,
they introduce large, difficult to correct error terms in the objective function, and training does not
converge. During testing, however, we still apply the fully-convolutional RPN to the entire image.
This may generate cross-boundary proposal boxes, which we clip to the image boundary.
Some RPN proposals highly overlap with each other. To reduce redundancy, we adopt non-
maximum suppression (NMS) on the proposal regions based on their cls scores. We fix the IoU
threshold for NMS at 0.7, which leaves us about 2k proposal regions per image. As we will show,
NMS does not harm the ultimate detection accuracy, but substantially reduces the number of pro-
posals. After NMS, we use the top-N ranked proposal regions for detection. In the following, we
train Fast R-CNN using 2k RPN proposals, but evaluate different numbers of proposals at test-time.

4 Experiments

We comprehensively evaluate our method on the PASCAL VOC 2007 detection benchmark [4].
This dataset consists of about 5k trainval images and 5k test images over 20 object categories. We
also provide results in the PASCAL VOC 2012 benchmark for a few models. For the ImageNet
pre-trained network, we use the “fast” version of ZF net [22] that has 5 conv layers and 3 fc layers,
and the public VGG-16 model5 [18] that has 13 conv layers and 3 fc layers. We primarily evalu-
ate detection mean Average Precision (mAP), because this is the actual metric for object detection
(rather than focusing on object proposal proxy metrics).
Table 1 (top) shows Fast R-CNN results when trained and tested using various region proposal
methods. These results use the ZF net. For Selective Search (SS) [21], we generate about 2k SS
proposals by the “fast” mode. For EdgeBoxes (EB) [23], we generate the proposals by the default
EB setting tuned for 0.7 IoU. SS has an mAP of 58.7% and EB has an mAP of 58.6%. RPN with Fast
R-CNN achieves competitive results, with an mAP of 59.9% while using only 300 proposals. Using
RPN for proposals and limiting Fast R-CNN to the top 300 regions yields a much faster detector
than using either SS or EB. Next, we consider several ablations of RPN and then show that proposal
quality improves when using the very deep VGG-16 network.
5
[Link]/˜vgg/research/very_deep/

5
Table 1: Detection results on PASCAL VOC 2007 test set (trained on VOC 2007 trainval). The
detectors are Fast R-CNN with ZF, but using various proposal methods for training and testing.
train-time region proposals test-time region proposals
method # boxes method # proposals mAP (%)
SS 2k SS 2k 58.7
EB 2k EB 2k 58.6
RPN+ZF, shared 2k RPN+ZF, shared 300 59.9
ablation experiments follow below
RPN+ZF, unshared 2k RPN+ZF, unshared 300 58.7
SS 2k RPN+ZF 100 55.1
SS 2k RPN+ZF 300 56.8
SS 2k RPN+ZF 1k 56.3
SS 2k RPN+ZF (no NMS) 6k 55.2
SS 2k RPN+ZF (no cls) 100 44.6
SS 2k RPN+ZF (no cls) 300 51.4
SS 2k RPN+ZF (no cls) 1k 55.8
SS 2k RPN+ZF (no reg) 300 52.1
SS 2k RPN+ZF (no reg) 1k 51.3
SS 2k RPN+VGG 300 59.2

Ablation Experiments. To investigate the behavior of RPNs as a proposal method, we conducted

several ablation studies. First, we show the effect of sharing conv layers between the RPN and Fast
R-CNN detection network. To do this, we stop after the second step in the 4-step training process.
Using separate networks reduces the result slightly to 58.7% (RPN+ZF, unshared, Table 1). We
observe that this is because in the third step when the detector-tuned features are used to fine-tune
the RPN, the proposal quality is improved.
Next, we disentangle the RPN’s influence on training the Fast R-CNN detection network. For this
purpose, we train a Fast R-CNN model by using the 2k SS proposals and ZF net. We fix this detector
and evaluate the detection mAP by changing the proposal regions used at test-time. In these ablation
experiments, the RPN does not share features with the detector.
Replacing SS with 300 RPN proposals at test-time leads to an mAP of 56.8%. The loss in mAP
is because of the inconsistency between the training/testing proposals. This result serves as the
baseline for the following comparisons.
Somewhat surprisingly, the RPN still leads to a competitive result (55.1%) when using the top-
ranked 100 proposals at test-time, indicating that the top-ranked RPN proposals are accurate. On
the other extreme, using the top-ranked 6k RPN proposals (without NMS) has a comparable mAP
(55.2%), suggesting NMS does not harm the detection mAP and may reduce false alarms.
Next, we separately investigate the roles of RPN’s cls and reg outputs by turning off either of them
at test-time. When the cls layer is removed at test-time (thus no NMS/ranking is used), we randomly
sample N proposals from the unscored regions. The mAP is nearly unchanged with N = 1k
(55.8%), but degrades considerably to 44.6% when N = 100. This shows that the cls scores account
for the accuracy of the highest ranked proposals.
On the other hand, when the reg layer is removed at test-time (so the proposals become anchor
boxes), the mAP drops to 52.1%. This suggests that the high-quality proposals are mainly due to
regressed positions. The anchor boxes alone are not sufficient for accurate detection.
We also evaluate the effects of more powerful networks on the proposal quality of RPN alone.
We use VGG-16 to train the RPN, and still use the above detector of SS+ZF. The mAP improves
from 56.8% (using RPN+ZF) to 59.2% (using RPN+VGG). This is a promising result, because it
suggests that the proposal quality of RPN+VGG is better than that of RPN+ZF. Because proposals of
RPN+ZF are competitive with SS (both are 58.7% when consistently used for training and testing),
we may expect RPN+VGG to be better than SS. The following experiments justify this hypothesis.

6
Table 2: Detection results on PASCAL VOC 2007 test set. The detector is Fast R-CNN and VGG-
16. Training data: “07”: VOC 2007 trainval, “07+12”: union set of VOC 2007 trainval and VOC
2012 trainval. For RPN, the train-time proposals for Fast R-CNN are 2k.
method # proposals data mAP (%) time (ms)
SS 2k 07 66.9 1830
SS 2k 07+12 70.0 1830
RPN+VGG, unshared 300 07 68.5 342
RPN+VGG, shared 300 07 69.9 196
RPN+VGG, shared 300 07+12 73.2 196

Table 3: Detection results on PASCAL VOC 2012 test set. The detector is Fast R-CNN and VGG-
16. Training data: “07”: VOC 2007 trainval, “07++12”: union set of VOC 2007 trainval+test
and VOC 2012 trainval. For RPN, the train-time proposals for Fast R-CNN are 2k. † : http://
[Link]/anonymous/[Link]. ‡ : [Link]
anonymous/[Link]
method # proposals data mAP (%)
SS 2k 12 65.7
SS 2k 07++12 68.4
RPN+VGG, shared† 300 12 67.0
RPN+VGG, shared‡ 300 07++12 70.4

Table 4: Timing (ms) on an Nvidia K40 GPU evaluated in Caffe [10], except SS proposal is evalu-
ated in a CPU. The region-wise computation includes NMS, pooling, fc, softmax, and others.
model system conv proposal region-wise total rate
VGG SS + Fast R-CNN 146 1510 174 1830 0.5 fps
VGG RPN + Fast R-CNN 146 10 40 196 5 fps
ZF RPN + Fast R-CNN 37 3 28 68 15 fps

Detection Accuracy and Running Time of VGG-16. Table 2 shows the results of VGG-16 for
both proposal and detection. The Fast R-CNN baseline of SS and VGG-16 has an mAP of 66.9%
[5]. Using RPN+VGG, the Fast R-CNN result is 68.5% for unshared features, 1.6% higher than
the SS baseline. As shown above, this is because the proposals generated by RPN+VGG are more
accurate than SS. Unlike SS that is pre-defined, the RPN is actively trained and benefits from better
networks. For the feature-shared variant, the result is 69.9%—better than the strong SS baseline,
yet with nearly cost-free proposals. We further train the RPN and detection network on the union
set of PASCAL VOC 2007 trainval and 2012 trainval, following [5]. The mAP is 73.2%, which is
3.2% higher than the SS counterpart. On the PASCAL VOC 2012 test set (Table 3), our method
has an mAP of 70.4% trained on the union set of VOC 2007 trainval+test and VOC 2012 trainval,
following [5]. This is 2.0% higher than the SS counterpart.
In Table 4 we summarize the running time of the entire object detection system. SS takes 1-2
seconds depending on content (on average 1.51s), and Fast R-CNN with VGG-16 takes 320ms on
2k SS proposals (or 223ms if using SVD on fc layers [5]). Our system with VGG-16 takes in total
196ms for both proposal and detection. With the conv features shared, the RPN alone only takes
10ms computing the additional layers. Our region-wise computation is also low, thanks to fewer
proposals (300). Our system has a rate of 15 fps with the ZF net.

Analysis of Recall-to-IoU. Next we compute the recall of proposals at different IoU ratios with
ground-truth boxes. It is noteworthy that the Recall-to-IoU metric is just loosely [9, 8, 1] related to
the ultimate detection accuracy. It is more appropriate to use this metric to diagnose the proposal
method than to evaluate it.
In Fig. 2, we show the results of using 300, 1k, and 2k proposals. We compare with SS and EB, and
the N proposals are the top-N ranked ones based on the confidence generated by these methods.
The plots show that the RPN method behaves gracefully when the number of proposals drops from
2k to 300. This explains why the RPN has a good ultimate detection mAP when using as few as 300

7
 ϯϬϬƉƌŽƉŽƐĂůƐ ϭϬϬϬƉƌŽƉŽƐĂůƐ ϮϬϬϬƉƌŽƉŽƐĂůƐ
ϭ ϭ ϭ

Ϭ͘ϴ Ϭ͘ϴ Ϭ͘ϴ

Ϭ͘ϲ Ϭ͘ϲ Ϭ͘ϲ

ZĞĐĂůů ^^ ^^ ^^
Ϭ͘ϰ Ϭ͘ϰ Ϭ͘ϰ
  
Ϭ͘Ϯ ZWE& Ϭ͘Ϯ ZWE& Ϭ͘Ϯ ZWE&
ZWEs'' ZWEs'' ZWEs''
Ϭ Ϭ Ϭ
Ϭ͘ϱ Ϭ͘ϲ Ϭ͘ϳ Ϭ͘ϴ Ϭ͘ϵ ϭ Ϭ͘ϱ Ϭ͘ϲ Ϭ͘ϳ Ϭ͘ϴ Ϭ͘ϵ ϭ Ϭ͘ϱ Ϭ͘ϲ Ϭ͘ϳ Ϭ͘ϴ Ϭ͘ϵ ϭ
/Žh /Žh /Žh
Figure 2: Recall vs. IoU overlap ratio on the PASCAL VOC 2007 test set.

Table 5: One-Stage Detection vs. Two-Stage Proposal + Detection. Detection results are on the
PASCAL VOC 2007 test set using the ZF model and Fast R-CNN. RPN uses unshared features.
regions detector mAP (%)
Two-Stage RPN + ZF, unshared 300 Fast R-CNN + ZF, 1 scale 58.7
One-Stage dense, 3 scales, 3 asp. ratios 20k Fast R-CNN + ZF, 1 scale 53.8
One-Stage dense, 3 scales, 3 asp. ratios 20k Fast R-CNN + ZF, 5 scales 53.9

proposals. As we analyzed before, this property is mainly attributed to the cls term of the RPN. The
recall of SS and EB drops more quickly than RPN when the proposals are fewer.
We also notice that for high IoU ratios (e.g., > 0.8), the recall of RPN and EB is lower than that of
SS, but the ultimate detection mAP is not impacted. This also suggests that the Recall-to-IoU curve
is not tightly related to the detection mAP.

One-Stage Detection vs. Two-Stage Proposal + Detection. The OverFeat paper [17] proposes a
detection method that uses regressors and classifiers on sliding windows over conv feature maps.
OverFeat is a one-stage, class-specific detection pipeline, and ours is a two-stage cascade consisting
of class-agnostic proposals and class-specific detections. In OverFeat, the region-wise features come
from a sliding window of one aspect ratio over a scale pyramid. These features are used to simulta-
neously determine the location and category of objects. In RPN, the features are from square (3×3)
sliding windows and predict proposals relative to anchors with different scales and aspect ratios.
Though both methods use sliding windows, the region proposal task is only the first stage of RPN
+ Fast R-CNN—the detector attends to the proposals to refine them. In the second stage of our cas-
cade, the region-wise features are adaptively pooled [7, 5] from proposal boxes that more faithfully
cover the features of the regions. We believe these features lead to more accurate detections.
To compare the one-stage and two-stage systems, we emulate the OverFeat system (and thus also
circumvent other differences of implementation details) by one-stage Fast R-CNN. In this system,
the “proposals” are dense sliding windows of 3 scales (128, 256, 512) and 3 aspect ratios (1:1, 1:2,
2:1). Fast R-CNN is trained to predict class-specific scores and regress box locations from these
sliding windows. Because the OverFeat system adopts multi-scale features, we also evaluate using
conv features extracted from 5 scales. We use those 5 scales as in [7, 5].
Table 5 compares the two-stage system and two variants of the one-stage system. Using the ZF
model, the one-stage system has an mAP of 53.9%. This is lower than the two-stage system (58.7%)
by 4.8%. This experiment justifies the effectiveness of cascaded region proposals and object detec-
tion. Note that the one-stage system is also slower as it has considerably more proposals to process.

5 Conclusion
We have presented Region Proposal Networks (RPNs) for efficient and accurate region proposal
generation. By sharing convolutional features with the down-stream detection network, the region
proposal step is nearly cost-free. Our method enables a unified, deep-learning-based object detection
system to run at 5-15 fps. The learned RPN also improves region proposal quality and thus the
overall object detection accuracy.

8
Table 6: Results on PASCAL VOC 2007 test set with Fast R-CNN detectors and VGG-16. For RPN,
the train-time proposals for Fast R-CNN are 2k. RPN∗ denotes the unsharing feature version.
method # box data mAP areo bike bird boat bottle bus car cat chair cow table dog horse mbike person plant sheep sofa train tv

SS 2k 07 66.9 74.5 78.3 69.2 53.2 36.6 77.3 78.2 82.0 40.7 72.7 67.9 79.6 79.2 73.0 69.0 30.1 65.4 70.2 75.8 65.8
SS 2k 07+12 70.0 77.0 78.1 69.3 59.4 38.3 81.6 78.6 86.7 42.8 78.8 68.9 84.7 82.0 76.6 69.9 31.8 70.1 74.8 80.4 70.4
RPN∗ 300 07 68.5 74.1 77.2 67.7 53.9 51.0 75.1 79.2 78.9 50.7 78.0 61.1 79.1 81.9 72.2 75.9 37.2 71.4 62.5 77.4 66.4
RPN 300 07 69.9 70.0 80.6 70.1 57.3 49.9 78.2 80.4 82.0 52.2 75.3 67.2 80.3 79.8 75.0 76.3 39.1 68.3 67.3 81.1 67.6
RPN 300 07+12 73.2 76.5 79.0 70.9 65.5 52.1 83.1 84.7 86.4 52.0 81.9 65.7 84.8 84.6 77.5 76.7 38.8 73.6 73.9 83.0 72.6

Table 7: Results on PASCAL VOC 2012 test set with Fast R-CNN detectors and VGG-16. For RPN,
the train-time proposals for Fast R-CNN are 2k.
method # box data mAP areo bike bird boat bottle bus car cat chair cow table dog horse mbike person plant sheep sofa train tv

SS 2k 12 65.7 80.3 74.7 66.9 46.9 37.7 73.9 68.6 87.7 41.7 71.1 51.1 86.0 77.8 79.8 69.8 32.1 65.5 63.8 76.4 61.7
SS 2k 07++12 68.4 82.3 78.4 70.8 52.3 38.7 77.8 71.6 89.3 44.2 73.0 55.0 87.5 80.5 80.8 72.0 35.1 68.3 65.7 80.4 64.2
RPN 300 12 67.0 82.3 76.4 71.0 48.4 45.2 72.1 72.3 87.3 42.2 73.7 50.0 86.8 78.7 78.4 77.4 34.5 70.1 57.1 77.1 58.9
RPN 300 07++12 70.4 84.9 79.8 74.3 53.9 49.8 77.5 75.9 88.5 45.6 77.1 55.3 86.9 81.7 80.9 79.6 40.1 72.6 60.9 81.2 61.5

References
[1] N. Chavali, H. Agrawal, A. Mahendru, and D. Batra. Object-Proposal Evaluation Protocol is ’Gameable’.
arXiv: 1505.05836, 2015.
[2] J. Dai, K. He, and J. Sun. Convolutional feature masking for joint object and stuff segmentation. In
CVPR, 2015.
[3] D. Erhan, C. Szegedy, A. Toshev, and D. Anguelov. Scalable object detection using deep neural networks.
In CVPR, 2014.
[4] M. Everingham, L. Van Gool, C. K. I. Williams, J. Winn, and A. Zisserman. The PASCAL Visual Object
Classes Challenge 2007 (VOC2007) Results, 2007.
[5] R. Girshick. Fast R-CNN. arXiv:1504.08083, 2015.
[6] R. Girshick, J. Donahue, T. Darrell, and J. Malik. Rich feature hierarchies for accurate object detection
and semantic segmentation. In CVPR, 2014.
[7] K. He, X. Zhang, S. Ren, and J. Sun. Spatial pyramid pooling in deep convolutional networks for visual
recognition. In ECCV. 2014.
[8] J. Hosang, R. Benenson, P. Dollár, and B. Schiele. What makes for effective detection proposals?
arXiv:1502.05082, 2015.
[9] J. Hosang, R. Benenson, and B. Schiele. How good are detection proposals, really? In BMVC, 2014.
[10] Y. Jia, E. Shelhamer, J. Donahue, S. Karayev, J. Long, R. Girshick, S. Guadarrama, and T. Darrell. Caffe:
Convolutional architecture for fast feature embedding. arXiv:1408.5093, 2014.
[11] A. Krizhevsky, I. Sutskever, and G. Hinton. Imagenet classification with deep convolutional neural net-
works. In NIPS, 2012.
[12] Y. LeCun, B. Boser, J. S. Denker, D. Henderson, R. E. Howard, W. Hubbard, and L. D. Jackel. Backprop-
agation applied to handwritten zip code recognition. Neural computation, 1989.
[13] J. Long, E. Shelhamer, and T. Darrell. Fully convolutional networks for semantic segmentation. In CVPR,
2015.
[14] V. Nair and G. E. Hinton. Rectified linear units improve restricted boltzmann machines. In ICML, 2010.
[15] S. Ren, K. He, R. Girshick, X. Zhang, and J. Sun. Object detection networks on convolutional feature
maps. arXiv:1504.06066, 2015.
[16] O. Russakovsky, J. Deng, H. Su, J. Krause, S. Satheesh, S. Ma, Z. Huang, A. Karpathy, A. Khosla,
M. Bernstein, A. C. Berg, and L. Fei-Fei. ImageNet Large Scale Visual Recognition Challenge.
arXiv:1409.0575, 2014.
[17] P. Sermanet, D. Eigen, X. Zhang, M. Mathieu, R. Fergus, and Y. LeCun. Overfeat: Integrated recognition,
localization and detection using convolutional networks. In ICLR, 2014.
[18] K. Simonyan and A. Zisserman. Very deep convolutional networks for large-scale image recognition. In
ICLR, 2015.
[19] C. Szegedy, S. Reed, D. Erhan, and D. Anguelov. Scalable, high-quality object detection.
arXiv:1412.1441v2, 2015.
[20] C. Szegedy, A. Toshev, and D. Erhan. Deep neural networks for object detection. In NIPS, 2013.
[21] J. R. Uijlings, K. E. van de Sande, T. Gevers, and A. W. Smeulders. Selective search for object recognition.
IJCV, 2013.
[22] M. D. Zeiler and R. Fergus. Visualizing and understanding convolutional neural networks. In ECCV,
2014.
[23] C. L. Zitnick and P. Dollár. Edge boxes: Locating object proposals from edges. In ECCV, 2014.

9
 person : 0.918 cow : 0.995
bird : 0.902

person : 0.988
person : 0.992
car : 0.745 person : 0.797 bird : 0.978
car : 0.955 horse : 0.991

bird : 0.972 cow : 0.998

bird : 0.941

bottle : 0.726

person : 0.964 person : 0.988 person : 0.986

car : 0.999 personperson
: 0.993 : 0.959
person : 0.976
person : 0.929 person : 0.994
person : 0.991 car : 0.997 car : 0.980

dog : 0.981

cow : 0.979 person : 0.998

person : 0.961 cow : 0.974
person : 0.958
cow : 0.979
bus : 0.999
person : 0.960 cow : 0.892
cow : 0.985
person : 0.985 person : 0.995
person : 0.996
person : 0.757 person : 0.994

dog : 0.697

cat : 0.998

person : 0.917
boat : 0.671
car : 1.000 boat : 0.895 boat : 0.749

boat : 0.877

person : 0.988

person : 0.995

bicycle :: 0.981
person : 0.994person 0.987
person : 0.930 person : 0.940
person : 0.893
bicycle : 0.972
bicycle : 0.977
boat : 0.992
person : 0.962
dog : 0.987
pottedplant : 0.951

bottle : 0.851
bottle : 0.962

boat : 0.693 diningtable : 0.791

boat : 0.846
person : 0.948
person : 0.972 person : 0.919

pottedplant : 0.728
car : 1.000 car : 0.880
car : 0.981
car : 0.982 chair : 0.630
boat : 0.995
boat : 0.948
diningtable : 0.862
bottle : 0.826

boat : 0.692
boat : 0.808

person : 0.975

aeroplane : 0.992 bird : 0.998

aeroplane : 0.986
sheep : 0.970

bird : 0.980

bird : 0.806 person : 0.670

horse : 0.984

aeroplane : 0.998

pottedplant : 0.820

chair : 0.984
diningtable : 0.997
pottedplant : 0.993 chair : 0.978
chair : 0.962
chair : 0.976
pottedplant : 0.715 car : 0.907 person : 0.987
person : 0.993

pottedplant : 0.940

pottedplant : 0.869

tvmonitor : 0.945
person : 0.983

aeroplane : 0.978 bird : 0.997

tvmonitor : 0.993

chair : 0.723
person : 0.968 chair : 0.982 tvmonitor : 0.993 person : 0.959
bottle : 0.789

person : 0.988
diningtable : 0.903 bottle : 0.858
chair : 0.852 bottle : 0.616bottle : 0.903
person : 0.897
person : 0.870

bottle : 0.884
bird : 0.727

Figure 3: Examples on the PASCAL VOC 2007 test set of ultimate detection results using RPN +
Fast R-CNN with shared features. The model is VGG-16 and the training data is 07+12 trainval.
Our method detects objects of a wide range of scales and aspect ratios. Each output box is associated
with a category label and a softmax score in [0, 1]. A score threshold of 0.6 is used to display these
images. The running time for obtaining these results is 196ms per image, including all steps.

10