Experimental results show that the proposed MEP outperforms other pipeline-based architecture and deep learning based approaches on the SG-MOD and KITTI-MOD datasets in most metrics.

A new dataset, Singapore motion object detection (SG-MOD) dataset is constructed in this work with much larger variations in urban environments. That consists of an object detector, a novel feature extraction algorithm to capture relative motion between objects and the background, as well as a motion predictor for representation learning with stacked autoencoder to determine motions. In this paper, we propose a motion embedding pipeline (MEP) architecture that utilizes interpretable optical flow and deep learning to solve object-level MOD problems. However, the interpretable optical flow feature has not been explored by object-level MOD approaches. Several deep learning based motion segmentation approaches have been reported based on the interpretable optical flow feature. Experiments on challenging real-world videos show promising performance in object-level motion detection.Īccurate motion object detection (MOD) using in-vehicle cameras in driving vehicles is a challenging task. vehicles and humans that are captured with different sizes, moving speed and image backgrounds by using a moving camera. The proposed technique has been evaluated over two types of widely studied objects, i.e. The inconsistency between the histogram of oriented optical flow (HOOF) of the object and its surrounding background is measured for the object-level motion detection. the optical flow of the image background surrounding the moving object of interest, which describes the object motion behavior better than other contexts such as camera’s GPS and direction. The CMD descriptor employs the contextual information, e.g. A context-aware motion descriptor (CMD) is designed based on the object’s moving speed and moving direction relative to that of the moving camera. This paper presents a novel technique that detects object-level motion from a freely moving camera using only two consecutive video frames. The object-level motion detection from moving cameras is a difficult problem due to the dual motion introduced by the mixture of the camera motion and the object motion. It is important for a moving observer to be able to identify his/her surrounding objects and determine whether these objects are moving or stationary, which is called object-level motion detection.

Web camera motion detection ware