Boundary-Aware Instance Segmentation

By Zeeshan Hayder, Xuming He, Mathieu Salzmann. 2016

160 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

Hybrid Task Cascade for Instance Segmentation (2019, methodology): “Other approaches include bridging category-leval and instance-level segmentation [42], learning a boundary-aware mask representation [17], and employing a sequence of neural networks to deal with different sub-grouping problems [27].”
Path Aggregation Network for Instance Segmentation (2018, background): “[20] merged feature pooling to network for faster speed.”
MaskLab: Instance Segmentation by Refining Object Detection with Semantic and Direction Features (2017, background): “[30] improve MNC by recovering the mask boundary error resulted from box prediction.”
Playing for Benchmarks (2017, methodology): “We benchmark the performance of Multi-task Network Cascades (MNC) [12] and of Boundary-aware In-stance Segmentation (BAIS) [25].”
Image Segmentation Using Deep Learning: A Survey (2020, methodology): “Other instance segmentation models have been developed based on R-CNN, such as those developed for mask proposals, including R-FCN [68], DeepMask [69], PolarMask [70], boundary-aware instance segmentation [71], and CenterMask [72].”
Instance Segmentation by Jointly Optimizing Spatial Embeddings and Clustering Bandwidth (2019, background): “d is very good at detecting small objects and often predicts more correct objects than annotated in the ground-truth. Method AP AP 50 FPS Deep Contours [27] 2.3 3.6 5 Box2Pix [26] 13.1 27.2 10.9 BAIS [8] 17.4 36.7 <1 Discriminate Loss [6] 17.5 35.9 5 DWT [3] 19.4 35.3 <3 Dynamic Net [2] 20.0 3”
ShapeMask: Learning to Segment Novel Objects by Refining Shape Priors (2019, methodology): “The detection-based approaches [17, 10, 18, 28, 6, 19, 38, 39] first detect the bounding box for each object instance and predict the segmentation mask inside the region cropped by the detected box.”
Semi-convolutional Operators for Instance Segmentation (2018, background): “[22] extends [10] to model the shape of objects.”
Box2Pix: Single-Shot Instance Segmentation by Assigning Pixels to Object Boxes (2018, background): “Boundary-aware Instance Segmentation [21] presents an extension for Faster R-CNN predicting per-pixel distances to proposal centers in order to allow some pixels outside of the predicted object box to still be assigned to the respective box.”
OccuSeg: Occupancy-Aware 3D Instance Segmentation (2020, methodology): “based methods [10, 6, 16, 15, 23, 44] firstly generate region proposals (predefined rectangles) that contain objects and further classify pixels inside each proposal as objects or background.”
Seamless Scene Segmentation (2019, background): “The work in [18] emphasizes on refining object boundaries for binary segmentation masks initially generated from bounding box proposals.”
Mask R-CNN (2020, unknown): “9 SAIS [45] fine - 17.”
YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information (2024, methodology): “In addition, some researchers have tried to develop more general objective functions, such as loss function [5, 45, 46, 50, 77, 78], label assignment [10, 12, 33, 67, 79] and auxiliary supervision [18, 20, 24, 28, 29, 51, 54, 68, 76].”
FGN: Fully Guided Network for Few-Shot Instance Segmentation (2020, background): “Existing CNN-based approaches to instance segmentation are mostly fully-supervised, which require abundant labeled data for model training [12, 24, 11].”
Semantic Edge Detection with Diverse Deep Supervision (2018, methodology): “…utilizing rich feature hierarchies from deep networks for SED as well as other high-level tasks such as semantic segmentation (Chen et al. 2016; Maninis et al. 2017), object detection (Ferrari et al. 2008; Maninis et al. 2017), and instance segmentation (Hayder et al. 2017; Kirillov et al. 2017).”
SegFix: Model-Agnostic Boundary Refinement for Segmentation (2020, methodology): “Distance/Direction Map for Segmentation: Some recent work [3,24,53] performed distance transform to compute distance maps for instance segmentation task.”