Problem Of Moving Object Detection Research Articles

Robot Programming from a Single Demonstration for High Precision Industrial Insertion

We propose a novel approach for robotic industrial insertion tasks using the Programming by Demonstration technique. Our method allows robots to learn a high-precision task by observing human demonstration once, without requiring any prior knowledge of the object. We introduce an Imitated-to-Finetuned approach that generates imitated approach trajectories by cloning the human hand's movements and then fine-tunes the goal position with a visual servoing approach. To identify features on the object used in visual servoing, we model object tracking as the moving object detection problem, separating each demonstration video frame into the moving foreground that includes the object and demonstrator's hand and the static background. Then a hand keypoints estimation function is used to remove the redundant features on the hand. The experiment shows that the proposed method can make robots learn precision industrial insertion tasks from a single human demonstration.

Outlier-Robust Moving Object and Background Decomposition via Structured ${\ell }_p$-Regularized Low-Rank Representation

Low-rank and sparse decomposition is a powerful tool in background extraction and moving object detection problems since the static background usually lies in a low dimensional subspace while the rapidly changing object maintains a sparse property. Conventional techniques for low-rank and sparse decomposition can handle noise-free or Gaussian noise scenarios. However, their performance significantly degrades in challenging scenarios contaminated by impulsive noise or outliers. In this paper, a new model with a separable noise component, which we call unstructured <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">l</i> <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><b>p</b></sub> -regularized low-rank representation (ULLR), is proposed. Instead of incorporating the Frobenius norm constraint on the noise component, the proposed ULLR incorporates flexible <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">l</i> <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><b>p</b></sub> -regularized schemes with 0 <; p ≤ 2 to handle varied scenarios. For the moving object detection problem, we further incorporate the local group structure and homogeneous group structure to encode the underlying object. Then, a new model called the structured <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">l</i> <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><b>p</b></sub> -regularized low-rank representation (SLLR) is proposed to enhance the detected foreground while suppressing the noise and dynamic background. Especially, when 0 <; p ≤ 1, the SLLR achieves robustness against outliers and hence handles challenging scenarios well. The resultant optimization problem is solved using the framework of alternating direction method of multipliers (ADMM). Each ADMM subproblem can be efficiently solved. More importantly, the solutions for the noise-based subproblem are rigorously derived with respect to different <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">l</i> <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><b>p</b></sub> -regularized schemes. Simulation results demonstrate the efficiency, accuracy, and robustness to outliers of the proposed algorithms.

A unified approach to moving object detection in 2D and 3D scenes

The detection of moving objects is important in many tasks. Previous approaches to this problem can be broadly divided into two classes: 2D algorithms which apply when the scene can be approximated by a flat surface and/or when the camera is only undergoing rotations and zooms, and 3D algorithms which work well only when significant depth variations are present in the scene and the camera is translating. We describe a unified approach to handling moving object detection in both 2D and 3D scenes, with a strategy to gracefully bridge the gap between those two extremes. Our approach is based on a stratification of the moving object detection problem into scenarios which gradually increase in their complexity. We present a set of techniques that match the above stratification. These techniques progressively increase in their complexity, ranging from 2D techniques to more complex 3D techniques. Moreover, the computations required for the solution to the problem at one complexity level become the initial processing step for the solution at the next complexity level. We illustrate these techniques using examples from real-image sequences.