In this dissertation, we address the problem of detecting humans and vehicles, tracking them in crowded scenes, and finally determining their activities in aerial video. Even though this is a well explored problem in the field of computer vision, many challenges still remain when one is presented with realistic data. These challenges include large camera motion, strong scene parallax, fast obje...

In real-world crowd counting applications, the crowd densities vary greatly in spatial and temporal domains. A detection based counting method will estimate crowds accurately in low density scenes, while its reliability in congested areas is downgraded. A regression based approach, on the other hand, captures the general density information in crowded regions. Without knowing the location of ea...

We present a robust and precise localization system that achieves centimeter-level localization accuracy in disparate city scenes. Our system adaptively uses information from complementary sensors such as GNSS, LiDAR, and IMU to achieve high localization accuracy and resilience in challenging scenes, such as urban downtown, highways, and tunnels. Rather than relying only on LiDAR intensity or 3...

Recent works have shown that, even with simple low level visual cues, complex behaviors can be extracted automatically from crowded scenes, e.g. those depicting public spaces recorded from video surveillance cameras. However, low level features as optical flow or foreground pixels are inherently noisy. In this paper we propose a novel unsupervised learning approach for the analysis of complex s...

Along with the widespread growth of surveillance cameras, computer vision algorithms have played a fundamental role in analyzing the large amount of videos. However, most of the current approaches in automatic video surveillance assume that the observed scene is not crowded, and is composed of easily perceptible components. These approaches are hard to be extended to more challenging videos of ...

• A two-stage framework to achieve quadrotor obstacle avoidance with monocular vision is proposed. dueling double deep recurrent Q network trained learn the policy. The employs unsupervised learning based depth estimation for perception. This paper proposes a novel learning-based realize autonomous vision. adopts architecture, consisting of sensing module and decision module. in an manner can e...

Abstract It is still a great challenge to detect pedestrians in crowded scenes. Recent works mainly focus on enhancing the feature representation of visible region. Although this strategy effective, it fails fully exploit appearance information contained full-body To end, we propose Dual-Mimic learning model, which consists occluded unoccluded (Occ-Unocc) and (Full-Vis) mimic branches. Both bra...