Robust Visual Object Tracking with Two-Stream
Residual Convolutional Networks
Ning Zhang
∗
, Jingen Liu
∗
, Ke Wang
†
, Dan Zeng
‡
and Tao Mei
§
∗
JD AI Research, Mountain View, USA.
†
Migu Culture & Technology, Beijing, China
‡
Shanghai University, Shanghai, China
§
JD AI Research, Beijing, China
∗
{ning.zhang,jingen.liu}@jd.com,
†
wangke@migu.cn,
‡
dzeng@shu.edu.cn,
§
tmei@live.com
Abstract—The current deep learning based visual tracking
approaches have been very successful by learning the target
classification and/or estimation model from a large amount
of supervised training data in offline mode. However, most
of them can still fail in tracking objects due to some more
challenging issues such as dense distractor objects, confusing
background, motion blurs, and so on. Inspired by the human
“visual tracking” capability which leverages motion cues to
distinguish the target from the background, we propose a Two-
Stream Residual Convolutional Network (TS-RCN) for visual
tracking, which successfully exploits both appearance and motion
features for model update. Our TS-RCN can be integrated with
existing deep learning based visual trackers. To further improve
the tracking performance, we adopt a “wider” residual network
ResNeXt as its feature extraction backbone. To the best of our
knowledge, TS-RCN is the first end-to-end trainable two-stream
visual tracking system, which makes full use of both appearance
and motion features of the target. We have extensively evaluated
the TS-RCN on most widely used benchmark datasets including
VOT2018, VOT2019, and GOT-10K. The experiment results have
successfully demonstrated that our two-stream model can greatly
outperform the appearance based tracker, and it also achieves
state-of-the-art performance. The tracking system can run at up
to 38.1 FPS.
I. INTRODUCTION
Generic visual object tracking predicts the location of a
class-agnostic object at every frame of a video sequence. It
is a highly challenging task due to its class-agnostic nature,
background distraction, illumination discrepancy, motion blur,
and many more [1], [2]. In general, a visual tracking system
needs to perform two tasks simultaneously: target classifica-
tion and bounding-box estimation [3]. The former task is to
coarsely identify the target object region in current frame from
the background, while the latter further estimates the precise
bounding-box (i.e., tracker state) of the target object.
Recently, researchers have made great progress in visual
object tracking by exploiting the effective power of deep
convolution networks. The Siamese tracking approaches [4]
leverage a large amount of supervised data to learn a more
general region similarity measurement in offline mode, which
enables tracking to be performed by searching image regions
most similar to the target template. Due to the lack of
background appearance (e.g., distractor objects ), however, the
Siamese approaches are inferior to deal with unseen objects
and distractor objects. To address these limits, Bhat et al. [5]
propose a discriminative learning architecture (DiMP) which
is able to fully exploit both target and background appearance.
Fig. 1: The illustration of limitations for appearance based
visual tracking. The first column shows the tracking results
where the Green, Blue, and Red bounding box represent
groundtruth, DiMP tracker, and our TS-RCN tracker, respec-
tively. The second column illustrates the HSV-color visualiza-
tion of the optical flow. In all three cases, the target optical
flow has a different pattern than that of its local background.
Row (a) shows dense similar objects (i.e. crabs) as distractors;
Row (b) shows confusing background textures as distractors
(i.e., the flying drone blends with background buildings); Row
(c) shows the target (i.e., soccer ball) has motion blurs. This
figure is best viewed in PDF format.
Although DiMP is trained to separate the background from
target, it may still fail when the background becomes more
confusing and challenging. As shown in Fig. 1 (a) and (b),
DiMP is not able to track the targets (blue bounding box) due
to same-class distractors (i.e., similar crabs) and confusing
background texture, respectively. Additionally, the tracking
can be disconnected if the current frame has motion blurs
on the target. For example, the soccer ball is blurred due to
high speed as shown in 1 (c). All the aforementioned issues
can happen in most deep learning based tracking approaches
arXiv:2005.06536v1 [cs.CV] 13 May 2020
下载后可阅读完整内容，剩余7页未读，立即下载
普通网友
- 粉丝: 1w+
- 资源: 81
我的内容管理
展开
最新资源
- WebLogic集群配置与管理实战指南
- AIX5.3上安装Weblogic 9.2详细步骤
- 面向对象编程模拟试题详解与解析
- Flex+FMS2.0中文教程：开发流媒体应用的实践指南
- PID调节深入解析：从入门到精通
- 数字水印技术：保护版权的新防线
- 8位数码管显示24小时制数字电子钟程序设计
- Mhdd免费版详细使用教程：硬盘检测与坏道屏蔽
- 操作系统期末复习指南：进程、线程与系统调用详解
- Cognos8性能优化指南：软件参数与报表设计调优
- Cognos8开发入门：从Transformer到ReportStudio
- Cisco 6509交换机配置全面指南
- C#入门：XML基础教程与实例解析
- Matlab振动分析详解：从单自由度到6自由度模型
- Eclipse JDT中的ASTParser详解与核心类介绍
- Java程序员必备资源网站大全
资源上传下载、课程学习等过程中有任何疑问或建议，欢迎提出宝贵意见哦~我们会及时处理！
点击此处反馈
