• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권11호
• /
• pp.5555-5567
• /
• 2018

State-of-art instance segmentation networks are successful at generating 2D segmentation mask for region proposals with highest classification score, yet 3D object segmentation task is limited to geocentric embedding or detector of Sliding Shapes. To this end, we propose an amodal 3D instance segmentation network called A3IS-CNN, which extends the detector of Deep Sliding Shapes to amodal 3D instance segmentation by adding a new branch of 3D ConvNet called A3IS-branch. The A3IS-branch which takes 3D amodal ROI as input and 3D semantic instances as output is a fully convolution network(FCN) sharing convolutional layers with existing 3d RPN which takes 3D scene as input and 3D amodal proposals as output. For two branches share computation with each other, our 3D instance segmentation network adds only a small overhead of 0.25 fps to Deep Sliding Shapes, trading off accurate detection and point-to-point segmentation of instances. Experiments show that our 3D instance segmentation network achieves at least 10% to 50% improvement over the state-of-art network in running time, and outperforms the state-of-art 3D detectors by at least 16.1 AP.

• International Journal of Internet, Broadcasting and Communication
• /
• 제16권1호
• /
• pp.280-286
• /
• 2024

In this paper, layered UNet with warmup and dropout tricks was used to segment teeth instantly by using data labeled for each individual tooth and increase performance of the result. The layered UNet proposed before showed very good performance in tooth segmentation without distinguishing tooth number. To do instance segmentation of teeth, we labeled teeth CBCT data according to tooth numbering system which is devised by FDI World Dental Federation notation. Colors for labeled teeth are like AI-Hub teeth dataset. Simulation results show that layered UNet does also segment very well for each tooth distinguishing tooth number by color. Layered UNet model using warmup trick was the best with IoU values of 0.80 and 0.77 for training, validation data. To increase the performance of instance segmentation of teeth, we need more labeled data later. The results of this paper can be used to develop medical software that requires tooth recognition, such as orthodontic treatment, wisdom tooth extraction, and implant surgery.

• 스마트미디어저널
• /
• 제9권2호
• /
• pp.9-15
• /
• 2020

본 논문에서는 새롭게 제안하는 Multi-Path Encoder-Decoder 의 구조를 바탕으로 두개의 가지로 구성된 심층신경망을 통해서 영상 이미지에서 물체를 하나의 객체 단위로 분할 검출하는 방법을 제안하였다. 각 가지는 중심점 검출 가지(Dot branch), 객체 분할 가지(Segmentation branch)라 하고 중심점 검출 가지는 이미지로부터 각 객체의 중심점을 찾는 역할을 수행하고, 객체 분할 가지는 각 객체의 영역을 이미지로부터 분할하는 역할을 수행한다. 실험에서는 CVPPP 식물 이미지의 나뭇잎을 각각 구분하도록 학습 하였으며 중심점 검출 가지는 각 나뭇잎의 중심점들을 찾아내고, 객체 분할 가지는 원본 이미지와 찾아낸 중심점 이미지를 통하여 각 중심점에 해당하는 나뭇잎의 픽셀 분할 영역을 최종적으로 예측하게 된다. 기존의 객체 분할에서는 다양한 크기, 위치의 앵커박스를 만들어서 많은 영역(N > 1k)의 물체를 확인해야하는 연산량 문제점 혹은 이미지에서 고정되지 않는 총 객체의 개수를 예측하기 어려웠던 문제가 있었다. 제안한 심층신경망에서는 중심점을 기반으로 객체를 찾아내는 효과적인 방법을 제안하였다.

• 대한임베디드공학회논문지
• /
• 제16권5호
• /
• pp.209-214
• /
• 2021

Panoptic segmentation, which is now widely used in computer vision such as medical image analysis, and autonomous driving, helps understanding an image with holistic view. It identifies each pixel by assigning a unique class ID, and an instance ID. Specifically, it can classify 'thing' from 'stuff', and provide pixel-wise results of semantic prediction and object detection. As a result, it can solve both semantic segmentation and instance segmentation tasks through a unified single model, producing two different contexts for two segmentation tasks. Semantic segmentation task focuses on how to obtain multi-scale features from large receptive field, without losing low-level features. On the other hand, instance segmentation task focuses on how to separate 'thing' from 'stuff' and how to produce the representation of detected objects. With the advances of both segmentation techniques, several panoptic segmentation models have been proposed. Many researchers try to solve discrepancy problems between results of two segmentation branches that can be caused on the boundary of the object. In this survey paper, we will introduce the concept of panoptic segmentation, categorize the existing method into two representative methods and explain how it is operated on two methods: top-down method and bottom-up method. Then, we will analyze the performance of various methods with experimental results.

• 로봇학회논문지
• /
• 제14권1호
• /
• pp.31-39
• /
• 2019

A robot usually adopts ANN (artificial neural network)-based object detection and instance segmentation algorithms to recognize objects but creating datasets for these algorithms requires high labeling costs because the dataset should be manually labeled. In order to lower the labeling cost, a new scheme is proposed that can automatically generate a training images and label them for specific objects. This scheme uses an instance segmentation algorithm trained to give the masks of unknown objects, so that they can be obtained in a simple environment. The RGB images of objects can be obtained by using these masks, and it is necessary to label the classes of objects through a human supervision. After obtaining object images, they are synthesized with various background images to create new images. Labeling the synthesized images is performed automatically using the masks and previously input object classes. In addition, human intervention is further reduced by using the robot arm to collect object images. The experiments show that the performance of instance segmentation trained through the proposed method is equivalent to that of the real dataset and that the time required to generate the dataset can be significantly reduced.

• Journal of Ubiquitous Convergence Technology
• /
• 제1권1호
• /
• pp.18-22
• /
• 2007

Multi-Instance Learning(MIL) performs well to deal with inherently ambiguity of images in multimedia retrieval. In this paper, an effective framework for Contented-Based Image Retrieval(CBIR) with MIL techniques is proposed, the effective mechanism is based on the image segmentation employing improved Mean Shift algorithm, and processes the segmentation results utilizing mathematical morphology, where the goal is to detect the semantic concepts contained in the query. Every sub-image detected is represented as a multiple features vector which is regarded as an instance. Each image is produced to a bag comprised of a flexible number of instances. And we apply a few number of MIL algorithms in this framework to perform the retrieval. Extensive experimental results illustrate the excellent performance in comparison with the existing methods of CBIR with MIL.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제12권5호
• /
• pp.217-228
• /
• 2023

비디오 개체 분할은 비디오를 구성하는 영상 프레임 각각에 대해 관심 개체 분할을 수행해야 할 뿐만 아니라, 해당 비디오를 구성하는 프레임 시퀀스 전체에 걸쳐 개체들에 대한 정확한 트래킹을 요구하기 때문에 난이도가 높은 기술이다. 특히 드라마 비디오에서 인물 개체 분할은 다양한 장소와 시간대에서 상호 작용하는 복수의 주요 등장인물들에 대한 정확한 트래킹을 요구하는 특징을 가지고 있다. 또한, 드라마 비디오 인물 개체분할은 주연 인물들과 조연 혹은 보조 출연 인물들 간의 등장 빈도에 상당한 차이가 있어 일종의 클래스 불균형 문제도 있다. 본 논문에서는 미생 드라마 비디오들을 토대로 구축한 인물 개체 분할 데이터 집합인 MHIS를 소개하고, 등장인물 클래스 간의 심각한 데이터 불균형 문제를 효과적으로 해결하기 위한 새로운 비디오 데이터 보강 기법인 CDVA를 제안한다. 기존의 비디오 데이터 보강 기법들과는 달리, 새로운 CDVA 보강 기법은 비디오들의 시-공간적 맥락을 충분히 고려해서 목표 인물이 삽입되어야 할 배경 클립 내의 위치를 결정함으로써, 보다 더 현실적인 보강 비디오들을 생성한다. 따라서 본 논문에서 제안하는 새로운 비디오 데이터 보강 기법인 CDVA는 비디오 개체 분할을 위한 심층 신경망 모델의 성능을 효과적으로 향상시킬 수 있다. 본 논문에서는 MHIS 데이터 집합을 이용한 다양한 정량 및 정성 실험들을 통해, 제안 비디오 데이터 보강 기법의 유용성과 효과를 입증한다.

• Journal of information and communication convergence engineering
• /
• 제21권1호
• /
• pp.17-23
• /
• 2023

Recently, several high-performance instance segmentation models have used the Mask R-CNN model as a baseline, which reached a historical peak in instance segmentation in 2017. There are numerous derived models using the Mask R-CNN model, and if the performance of Mask R-CNN is improved, the performance of the derived models is also anticipated to improve. The Mask R-CNN uses interpolation to adjust the image size, and the input differs depending on the interpolation method. Therefore, in this study, the performance change of Mask R-CNN was compared when various interpolation methods were applied to the transform layer to improve the performance of Mask R-CNN. To train and evaluate the models, this study utilized the PennFudan and Balloon datasets and the AP metric was used to evaluate model performance. As a result of the experiment, the derived Mask R-CNN model showed the best performance when bicubic interpolation was used in the transform layer.

• 한국건축시공학회:학술대회논문집
• /
• 한국건축시공학회 2020년도 가을 학술논문 발표대회
• /
• pp.55-56
• /
• 2020

The construction industry has the highest occupational fatality and injury rates related to accidents of any industry. Accordingly, safety managers closely monitor to prevent accidents in real-time by installing surveillance cameras at construction sites. However, due to human cognitive ability limitations, it is impossible to monitor many videos simultaneously, and the fatigue of the person monitoring surveillance cameras is also very high. Thus, to help safety managers monitor work and reduce the occupational accident rate, a study on object recognition in construction sites was conducted through surveillance cameras. In this study, we applied to the instance segmentation to identify the classification and location of objects and extract the size and shape of objects in construction sites. This research considers ways in which deep learning-based computer vision technology can be applied to safety management on a construction site.

• 토지주택연구
• /
• 제14권4호
• /
• pp.95-102
• /
• 2023

도시 지역에서 객체를 탐지하기 위해 드론 고해상도 영상에 기계 학습 알고리즘을 적용하는 다양한 연구가 진행되었다. 그러나 대부분의 차량 추출 연구는 인스턴스 세그멘테이션 대신 경계 박스로 차량을 탐지하여 차량의 방향이나 정확한 경계를 알 수 없다는 한계점이 있다. 인스턴스 세그멘테이션은 개별 개체를 훈련하기 위한 노동 집약적인 레이블링 작업을 필요로 하므로, 차량 추출을 위해 자동 무감독 인스턴스 세그멘테이션을 수행하는 방법에 대한 연구가 필요하다. 따라서 본 연구에서는 드론 영상의 차량 경계 박스에 대해 무감독 SVM 분류 기반의 차량 추출 기법을 제안하였다. 연구 결과, 차량을 89% 정확도로 추출할 수 있음을 확인하였으며 차량 내의 분광 특성이 크게 다른 경우에도 차량을 추출할 수 있음을 확인하였다.