Challenges in Video Based Object Detection in Maritime Scenario Using Computer Vision
13
Citation
0
Reference
10
Related Paper
Citation Trend
In this paper, we propose an object segmentation method to extract the same object from images that are captured from different views. Using matting method, the target object can be easily segmented with simple strokes given by a user to the desired object and background in the reference image. Then, homography is estimated after keypoint matching between the segmented object image and the target image, which is acquired from a different viewpoint, containing the same object. Using homography, the strokes are transformed to fit to a target image. Finally, the identical object is segmented from the target image using matting method with transformed strokes. The proposed method can be used to increase the accuracy of segmentation for extracting the same desired object from a target image.
Homography
Segmentation-based object categorization
Cite
Citations (0)
We study a method for unknown object detection based on impacting and keypoint tracking. In this method, a robot changes object positions by impacting to detect each of the objects individually from camera images before and after impacting. This detection is possible because keypoints of each object always move consistently by impacting, while those of the background do not move. A concave hull segmentation method called alpha-shape is used to model the objects. Picking experiments of several objects are demonstrated.
Tracking (education)
Cite
Citations (0)
3D object detection and pose estimation often requires a 3D object model, and even so, it is a difficult problem if the object is heavily occluded in a cluttered scene. In this paper, we introduce a novel approach for recognizing and localizing 3D objects based on their appearances through segmentation of 3D surfaces. The approach can identify multiple occluded objects in a scene, which may include different instances of the same object, and estimate the pose of each entire object even if the object can only be seen partially due to occlusion.
Cite
Citations (5)
This paper introduces a completely new moving object detection algorithm from a static background scene that uses colour images to contain shadows. Tracking of Object is based on analysis and detection of object motion, removal of background and shadow. Initially, an organisation is used and thought of as background data. The foreground and background object data is recognized based on reference frame of model when a brand new object enters the frame. The object shadow of background details is combined with the foreground object and makes the tracking of object is somewhat difficult. Structural based activities are used within the strategy to identify and remove the shadow. With the suggested algorithm, video sequences are captured and tested. Experimental outcomes are shown, showing the performance of the system.
Tracking (education)
Cite
Citations (0)
This paper proposes an object movement detection system covering large areas of a room by using multiple cameras. When object movement detection for whole of a room is performed, there are several challenging difficulties: sizes of objects on the camera images are small, non-objects such as humans also exist on the images, objects are sometimes difficult to detect in the specific viewpoints because of occlusion by humans or furniture or color similarity to near objects. In this work, we propose an object movement detection method by integrating multiple viewpoints via features extracted from “stable changes” on each viewpoint. To discriminate whether object or non-object, we focus on motion of changed regions. Experiment in a room environment shows the multiple view integration method with the color and position features improves recall rate of object detection performance.
Viewpoints
Similarity (geometry)
Position (finance)
Cite
Citations (0)
The object detection and tracking is the important steps of computer vision algorithm. The robust object detection is the challenge due to variations in the scenes. Another biggest challenge is to track the object in the occlusion conditions. Hence in this approach, the moving objects detection using TensorFlow object detection API. Further the location of the detected object is pass to the object tracking algorithm. A novel CNN based object tracking algorithm is used for robust object detection. The proposed approach is able to detect the object in different illumination and occlusion. The proposed approach achieved the accuracy of 90.88% on self generated image sequences.
Tracking (education)
Object-class detection
Cite
Citations (74)
The author proposes a method of feature-based camera-guided grasping of a known object by splitting up the 3D movement in several successive 1D or 2D movements. Object recognition was achieved by extracting the 3D features of the object from image sequences while a camera mounted on a robots hand was moving toward the object. After the recognition of the object, the camera approached the object by several camera-guided steps: motion in the xy-plane, rotations around the z-axis, and movements along the z-axis. These movements were controlled by data which were derived from the image features. Finally, the gripper had to do a fine motion to reach the correct position for the grasping of the object.< >
Feature (linguistics)
Position (finance)
Image plane
Cite
Citations (19)
We present a novel approach to weakly supervised object detection. Instead of annotated images, our method only requires two short videos to learn to detect a new object: 1) a video of a moving object and 2) one or more "negative" videos of the scene without the object. The key idea of our algorithm is to train the object detector to produce physically plausible object motion when applied to the first video and to not detect anything in the second video. With this approach, our method learns to locate objects without any object location annotations. Once the model is trained, it performs object detection on single images. We evaluate our method in three robotics settings that afford learning objects from motion: observing moving objects, watching demonstrations of object manipulation, and physically interacting with objects (see a video summary at https://youtu.be/BH0Hv3zZG_4).
3D single-object recognition
Object-class detection
Cite
Citations (1)
This study proposes a method to detect and mark the target object removed from the monitoring scene and the unknown object left in the monitoring scene. The present method uses the timeliness background to extract the foreground object and to mask the part that was unwanted. The foreground object was compared with the current frame, thus, the unreliable pixels were filtered out. By the identification of the center of mass (CoM) on foreground object, an object detection rule is developed to determine whether the foreground object is missing object or unattended object. In this paper, the present approach improves the problem with the high similarity of pixels between the foreground object and the background model. The experiment can be applied to any complex environment, both indoors and outdoors, such as the subway station, which is thronged with people. The experimental outcome, using the proposed method, can determine the missing and unattended object accurately and the unreasonable object is excluded in video surveillance system.
Object-class detection
Identification
Cite
Citations (10)
This paper describes a method for object detection from three-dimensional image using stereo vision. Each object has a different disparity distribution. In other words, an object's disparity distribution is discontinuous at the border between itself and another adjacent object. Focusing attention on this point, the image is divided into objects. Furthermore, objects having sharply-defined contour are detected using superpixels.
Computer stereo vision
Cite
Citations (0)