다시점 객체 공분할을 이용한 2D-3D 물체 자세 추정
2D-3D Pose Estimation using Multi-view Object Co-segmentation
1
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
We present a region-based approach for accurate pose estimation of small mechanical components. Our algorithm consists of two key phases: Multi-view object co-segmentation and pose estimation. In the first phase, we explain an automatic method to extract binary masks of a target object captured from multiple viewpoints. For initialization, we assume the target object is bounded by the convex volume of interest defined by a few user inputs. The co-segmented target object shares the same geometric representation in space, and has distinctive color models from those of the backgrounds. In the second phase, we retrieve a 3D model instance with correct upright orientation, and estimate a relative pose of the object observed from images. Our energy function, combining region and boundary terms for the proposed measures, maximizes the overlapping regions and boundaries between the multi-view co-segmentations and projected masks of the reference model. Based on high-quality co-segmentations consistent across all different viewpoints, our final results are accurate model indices and pose parameters of the extracted object. We demonstrate the effectiveness of the proposed method using various examples.
Abstract
We present a region-based approach for accurate pose estimation of small mechanical components. Our algorithm consists of two key phases: Multi-view object co-segmentation and pose estimation. In the first phase, we explain an automatic method to extract binary masks of a target object captured from multiple viewpoints. For initialization, we assume the target object is bounded by the convex volume of interest defined by a few user inputs. The co-segmented target object shares the same geometric representation in space, and has distinctive color models from those of the backgrounds. In the second phase, we retrieve a 3D model instance with correct upright orientation, and estimate a relative pose of the object observed from images. Our energy function, combining region and boundary terms for the proposed measures, maximizes the overlapping regions and boundaries between the multi-view co-segmentations and projected masks of the reference model. Based on high-quality co-segmentations consistent across all different viewpoints, our final results are accurate model indices and pose parameters of the extracted object. We demonstrate the effectiveness of the proposed method using various examples.
참고문헌 (29)
- 박동일, "구난 로봇용 양팔 머니퓰레이터 진동 해석 및 설계", 로봇학회 논문지, 11(4), pp.235~241 (2016), 한국로봇학회
- 박찬훈, "병렬로봇의 설계 공차가 궤적 정밀도에 미치는 영향 분석", 로봇학회 논문지, 11(4), pp.248~255 (2016), 한국로봇학회
- 도현민, "모듈러 로봇의 작업 적응성을 위한자가 재구성 제어 엔진", 로봇학회 논문지, 11(4), pp.270~276 (2016), 한국로봇학회
- 최태용, "산업용 양팔로봇을 이용한 공연 예술 구현 사례", 로봇학회 논문지, 11(4), pp.293~299 (2016), 한국로봇학회
- 김승훈, "비행로봇의 항공 영상 온라인 학습을 통한 지상로봇 검출 및 추적", 로봇학회 논문지, 10(2), pp.104~111 (2015), 한국로봇학회
- 허환, "공장환경에서 AGV를 위한 인공표식 기반의 포즈그래프 SLAM", 로봇학회 논문지, 10(2), pp.112~118 (2015), 한국로봇학회
- 최한수, "삼륜형 전방향 이동로봇을 위한 경로추종 및 방위제어", 로봇학회 논문지, 10(3), pp.154~161 (2015), 한국로봇학회
- 송승우, "기계식 중력보상 기반의 중량물 취급용 5자유도 로봇머니퓰레이터의 개발", 로봇학회 논문지, 11(4), pp.242~247 (2016), 한국로봇학회
- 안병태, "안정적인 실시간 얼굴 특징점 추적과 감정인식 응용", 로봇학회 논문지, 8(4), pp.266~272 (2013), 한국로봇학회
- 바쟝 정샬, "Mixing Collaborative and Hybrid Vision Devices for Robotic Applications", 로봇학회 논문지, 6(3), pp.210~219 (2011), 한국로봇학회
- S. -H. Kim, "Two phase approach for multi-view object extraction", Proceedings of International Conference on Image Processing (ICIP) (2011)
- S. -H. Kim, "Multi-view object extraction with fractional boundaries", IEEE Trans. on Image Processing (TIP), 25(8), pp.3639~3654 (2016)
- W. Lee, "Silhouette segmentation in multiple views", IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), 33(7), pp.1429~1441 (2011)
- A. Kowdle, "Multiple view object cosegmentation using appearance and stereo cues", European Conference on Computer Vision(ECCV) (2012)
- A. Djelouah, "Multi-view object segmentation in space and time", International Conference on Computer Vision(ICCV) (2013)
- A. Djelouah, "Sparse multi-view consistency for object segmentation", IEEE Trans. on Pattern Analysis and Machine Intelligence (PAMI), 37(9), pp.1890~1903 (2015)
- J. -Y. Guillemaut, "Joint multi-layer segmentation and reconstruction for free-viewpoint video applications", International Journal on Computer Vision (IJCV), 93(1), pp.73~100 (2011)
- Y. Boykov, "Interactive graph cuts for optimal boundary and region segmentation of objects in n-d images", ICCV (2001)
- C. Rother, "Grabcut : Interactive foreground extraction using iterated graph cuts", Proceedings of ACM SIGGRAPH, 23(3), pp.309~314 (2004)
- Y. -W. Tai, "Soft color segmentation and its applications", IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), 29(9), pp.1520~1537 (2007)
- "VisualSFM: A visual structure from motion system" (2011)[URL]
- Y. Boykov, "Efficient Approximate Energy Minimization via Graph Cuts", IEEE TPAMI, 20(12), pp.1222~1239 (2001)
- V. Kolmogorov, "What Energy Functions can be Minimized via Graph Cuts?", Pattern Analysis and Machine Intelligence (PAMI), 26(2), pp.147~159 (2004)
- Y. Boykov, "An Experimental Comparison of Min-Cut", Max-Flow Algorithms for Energy Minimization in Vision, 2004, pp.26(9), 1124~1137
- V. Prisacariu, "Pwp3d : Real-time segmentation and tracking of 3d objects", International Journal of Computer Vision (IJCV), 98(3), pp.335~354 (2012)
- A. Vedaldi, "VLFeat : An open and portable library of computer vision algorithms", Proceedings of the 18th ACM International Conf. on Multimedia (2010)
- H. Fu, "Upright orientation of man-made objects", ACM transactions on graphics (TOG), 27(3) (2008)
- "123D Design"[URL]
- "Physically Based Rendering Techniques"[URL]