![](https://csdnimg.cn/release/download_crawler_static/8701391/bg4.jpg)
an object such as the appearance of its background or of
other objects next to it.
Concerning the general role of the regions on object de-
tection, we briefly focus below on two of the reasons why
they are helpful on this task:
Discriminative feature diversification. Our hypothesis
is that having regions that render visible to their network-
components only a limited part of the object or only its
immediate surrounding forces each network-component to
discriminate image boxes solely based on the visual in-
formation that is apparent on them thus diversifying the
discriminative factors captured by our overall recognition
model. For example, if the border region depicted on fig-
ure 3i is replaced with one that includes its whole inner
content, then we would expect that the network-component
dedicated to it will not pay the desired attention on the vi-
sual content that is concentrated around the borders of an
object. We tested such a hypothesis by conducting an exper-
iment where we trained and tested two Multi-Region CNN
models consisted of two regions each. Model A included
the original box region and the border region depicted on
figure 3i that does not contain the central part of the ob-
ject. On model B, we replaced the rectangular ring with a
normal box of the same size as the outer box on the rect-
angular ring on figure 3i. Both of them were trained on
PASCAL VOC2007 [7] trainval set and tested on the test
set of the same challenge. Model A achieved 64.1% mAP
while Model B achieved 62.9% mAP which is 1.2 points
lower and validates our assumption.
Localization-aware representation. Moreover, we argue
that the multi-region architecture of our model as well as the
type of regions included address to a certain extent one of
the major problems on the detection task, which is the inac-
curate localization of object instances. We expect that hav-
ing multiple regions with network-components dedicated
on each of them, imposes soft constraints regarding the vi-
sual content allowed on each type of region for a given can-
didate detection box. We provide experimental support for
this argument by referring to sections 6.2 and 6.3.
3. Semantic Segmentation-Aware CNN Fea-
tures
To further diversify the features encoded by our repre-
sentation, we extend the Multi-Region CNN model so that
it can also incorporate semantic segmentation-aware CNN
features (see figure 4). We were motivated for this also
from the close connection between segmentation and detec-
tion, and the fact that segmentation related cues have been
empirically known to help object detection [6, 11, 22]. In
the context of our multi-region CNN network, the incorpo-
ration of the semantic segmentation-aware features is done
by properly adapting the two main modules of the network,
i.e., the activation-maps module and the region-adaptation
(a) Original box (b) Half left
(c) Half right (d) Half up
(e) Half bottom (f) Central Region
(g) Central Region (h) Border Region
(i) Border Region (j) Context. Region
Figure 3: Illustration of the regions components used.
modules (see architecture in figure 4). We hereafter refer to
the resulting modules as:
• Activation maps module for semantic segmentation-
aware features.
• Region adaptation module for semantic segmentation-
aware features.
It is important to note that the modules for the semantic
segmentation-aware features are trained without the use of
4