KAIST · 컴퓨터과학
Thang Vu 교수의 연구실은 3D 시각 인식 분야에서 핵심적인 기여를 하고 있으며, 특히 3D 인스턴스 세그멘테이션과 객체 검출의 정확도 향상을 위한 혁신적 아키텍처 개발에 집중하고 있습니다. 소프트 그룹핑 기반의 세그멘테이션(SoftGroup)과 샘플 일致성 네트워크(SCNet), 캐스케이드 RPN 등은 예측 불확실성 감소와 훈련-추론 간 IoU 분포 불일치 문제를 체계적으로 해결함으로써 성능을 극대화합니다. 연구는 정확성과 확장성의 균형을 추구하며, 실시간 적용 가능한 고성능 3D 인식 기술의 기반을 마련하고 있습니다.
표시된 성과는 수집된 데이터 기준으로 산출되며, 일부 차이가 있을 수 있습니다.
Existing state-of-the-art 3D instance segmentation methods perform semantic segmentation followed by grouping. The hard predictions are made when performing semantic segmentation such that each point is associated with a single class. However, the errors stemming from hard decision propagate into grouping that results in (1) low overlaps between the predicted instance with the ground truth and (2) substantial false positives. To address the aforementioned problems, this paper proposes a 3D insta
Cascaded architectures have brought significant performance improvement in object detection and instance segmentation. However, there are lingering issues regarding the disparity in the Intersection-over-Union (IoU) distribution of the samples between training and inference. This disparity can potentially exacerbate detection accuracy. This paper proposes an architecture referred to as Sample Consistency Network (SCNet) to ensure that the IoU distribution of the samples at training time is close
This paper considers an architecture referred to as Cascade Region Proposal\nNetwork (Cascade RPN) for improving the region-proposal quality and detection\nperformance by \\textit{systematically} addressing the limitation of the\nconventional RPN that \\textit{heuristically defines} the anchors and\n\\textit{aligns} the features to the anchors. First, instead of using multiple\nanchors with predefined scales and aspect ratios, Cascade RPN relies on a\n\\textit{single anchor} per location and per
This paper considers an architecture referred to as Cascade Region Proposal Network (Cascade RPN) for improving the region-proposal quality and detection performance by \textit{systematically} addressing the limitation of the conventional RPN that \textit{heuristically defines} the anchors and \textit{aligns} the features to the anchors. First, instead of using multiple anchors with predefined scales and aspect ratios, Cascade RPN relies on a \textit{single anchor} per location and performs mult
This paper considers a network referred to as SoftGroup for accurate and scalable 3D instance segmentation. Existing state-of-the-art methods produce hard semantic predictions followed by grouping instance segmentation results. Unfortunately, errors stemming from hard decisions propagate into the grouping, resulting in poor overlap between predicted instances and ground truth and substantial false positives. To address the abovementioned problems, SoftGroup allows each point to be associated wit
Cascaded architectures have brought significant performance improvement in object detection and instance segmentation. However, there are lingering issues regarding the disparity in the Intersection-over-Union (IoU) distribution of the samples between training and inference. This disparity can potentially exacerbate detection accuracy. This paper proposes an architecture referred to as Sample Consistency Network (SCNet) to ensure that the IoU distribution of the samples at training time is close
A bounding box commonly serves as the proxy for 2D object detection. However, extending this practice to 3D detection raises sensitivity to localization error. This problem is acute on flat objects since small localization error may lead to low overlaps between the prediction and ground truth. To address this problem, this paper proposes Sphere Region Proposal Network (SphereRPN) which detects objects by learning spheres as opposed to bounding boxes. We demonstrate that spherical proposals are m
