SOLOv2: Dynamic and Fast Instance Segmentation

By Xinlong Wang, Rufeng Zhang, Tao Kong, Lei Li, Chunhua Shen. 2020

1,141 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Instance Segmentation for Autonomous Log Grasping in Forestry Operations (2022, background): “The SOLOv2 network [13] instead deals directly with instance segmentation, where each square of a grid is responsible for generating a segmentation mask and a class label, without generating bounding boxes.”
Masked-attention Mask Transformer for Universal Image Segmentation (2021, methodology): “Follow-up methods either focus on detecting more precise bounding boxes [4, 6], or ﬁnding new ways to generate a dynamic number of masks, e.g ., using dynamic kernels [3, 49, 56] or clustering algorithms [11, 29].”
K-Net: Towards Unified Image Segmentation (2021, methodology): “These methods all rely on NMS to eliminate duplicated instance masks, which hinders end-to-end training.”
Reviving Iterative Training with Mask Guidance for Interactive Segmentation (2021, background): “These datasets stimulated the development of state-of-the-art instance segmentation algorithms [42, 43, 44, 45], providing a large variety of labeled classes with a sufficient number of examples.”
MaX-DeepLab: End-to-End Panoptic Segmentation with Mask Transformers (2020, methodology): “Our mask prediction head is inspired by CondInst [85] and SOLOv2 [92], which extend dynamic convolution [43, 99] to instance segmentation.”
Segmenter: Transformer for Semantic Segmentation (2021, methodology): “Our mask transformer is inspired by DETR [7], MaxDeepLab [52] and SOLO-v2 [55] which introduce object embeddings [7] to produce instance masks [52, 55].”
Lite-HRNet: A Lightweight High-Resolution Network (2021, methodology): “CondINS [40] and SOLOV2 [43] apply this design to the instance segmentation task, generating the parameters of the mask sub-network for each instance.”
ASF-YOLO: A Novel YOLO Model with Attentional Scale Sequence Fusion for Cell Instance Segmentation (2023, methodology): “The two-stage R-CNN series [1 – 3] and its one-stage variants [4,5] are classical CNN-based frameworks that have been used for segmentation tasks.”
BoxInst: High-Performance Instance Segmentation with Box Annotations (2020, background): “The performance of instance segmentation has been significantly advanced by a number of recent successful methods [5, 6, 11, 15, 28, 30, 31].”
RSPrompter: Learning to Prompt for Remote Sensing Instance Segmentation Based on Visual Foundation Model (2023, methodology): “We also take into account single-stage methods, such as SOLOv2 [89], CondInst [28], BoxInst [91], and Mask2Former [30].”
Sparse R-CNN: End-to-End Object Detection with Learnable Proposals (2020, background): “Moreover, NMS post-processing [1, 39] is needed to remove redundant predictions during inference time.”
CondLaneNet: a Top-to-down Lane Detection Framework Based on Conditional Convolution (2021, methodology): “Besides, we regress the dynamic kernel parameters by predicting a parameter map following CondInst [35] and SOLOv2 [38].”