Wearables and robots are getting smarter at recognizing objects, following commands, and navigating spaces—but they still struggle with something humans ...

Abstract: As an important area in computer vision, object tracking has formed two separate communities that respectively study Single Object Tracking (SOT) and Multiple Object Tracking (MOT). However, ...

In summer 2025, I had the opportunity to conduct research in the University of New Hampshire Vision Lab through the Research Experience and Apprenticeship Program (REAP). This lab uses a virtual ...

This project implements a comprehensive Computer Vision MLOps pipeline for aerial object analysis, specifically designed to classify and detect birds vs drones in aerial imagery. The system combines: ...

ABSTRACT: The DOFBOT Robotic Hand, powered by the NVIDIA Jetson Nano processor, represents a cutting-edge fusion of computer vision and robotic automation for precision agriculture. This research ...

What if you could teach a computer to recognize a zebra without ever showing it one? Imagine a world where object detection isn’t bound by the limits of endless training data or high-powered hardware.

advanced-computer-vision-framework/ ├── src/ │ ├── cuda/ # Custom CUDA kernels │ ├── tracking/ # Multi-object tracking │ ├── features/ # Feature extraction │ ├── layers/ # Custom neural network layers ...

Our proposed HOMATracker comprises the following steps: (A) Object detection: A detector is used to locate the objects in each frame; (B) Instance-level appearance feature extraction: We involve a ...

This study introduces Popnet, a deep learning model for forecasting 1 km-gridded populations, integrating U-Net, ConvLSTM, a Spatial Autocorrelation module and deep ensemble methods. Using spatial ...