arXiv:2009.04650
공식 코드 · Papers with Code
Towards Fine-grained Large Object Segmentation 1st Place Solution to 3D AI Challenge 2020 -- Instance Segmentation Track
원문 초록
This technical report introduces our solutions of Team 'FineGrainedSeg' for Instance Segmentation track in 3D AI Challenge 2020. In order to handle extremely large objects in 3D-FUTURE, we adopt PointRend as our basic framework, which outputs more fine-grained masks compared to HTC and SOLOv2. Our final submission is an ensemble of 5 PointRend models, which achieves the 1st place on both validation and test leaderboards. The code is available at https://github.com/zehuichen123/3DFuture_ins_seg.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
No time to train! Training-Free Reference-Based Instance Segmentation
NOCTIS: Novel Object Cyclic Threshold based Instance Segmentation
Prohibited Items Segmentation via Occlusion-aware Bilayer Modeling
The Four Color Theorem for Cell Instance Segmentation
OpenMaskDINO3D : Reasoning 3D Segmentation via Large Language Model
SCORE: Scene Context Matters in Open-Vocabulary Remote Sensing Instance Segmentation
토론
이 논문과 연결된 포럼 글이 아직 없습니다.