Interactive Image Segmentation via Backpropagating Refinement Scheme

By Won-Dong Jang, Chang-Su Kim. 2019

185 citations recorded by the source corpus.

An interactive image segmentation algorithm, which accepts user-annotations about a target object and the background, is proposed in this work. We convert user-annotations into interaction maps by measuring distances of each pixel to the annotated locations. Then, we perform the forward pass in a convolutional neural network, which outputs an initial segmentation map. However, the user-annotated locations can be mislabeled in the initial result. Therefore, we develop the backpropagating refinement scheme (BRS), which corrects the mislabeled pixels. Experimental results demonstrate that the proposed algorithm outperforms the conventional algorithms on four challenging datasets. Furthermore, we demonstrate the generality and applicability of BRS in other computer vision tasks, by transforming existing convolutional neural networks into user-interactive ones.

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

FocalClick: Towards Practical Interactive Image Segmentation (2022, methodology): “BRS [18] first introduces online optimization, which enables the model to update during annotation.”
Reviving Iterative Training with Mask Guidance for Interactive Segmentation (2021, methodology): “Recent works on click-based interactive segmentation focus on inference-time optimization procedures [1, 2].”
Getting to 99% Accuracy in Interactive Segmentation (2020, methodology): “60 BRS [14] 3.”
Conditional Diffusion for Interactive Segmentation (2021, methodology): “BRS [11] uses backward propagation to finetune the guidance map in an online manner. f-BRS [27] refines the intermediate feature to get more precise masks with faster speed compared with BRS.”
FocusCut: Diving into a Focus View in Interactive Segmentation (2022, methodology): “Method NoF20@90 NoF100@90 NoC100@90 BRS [20] 77 51 20.”
Interactive Object Segmentation With Inside-Outside Guidance (2022, methodology): “BRS [50] adopts back-propagation reﬁnement strategy to correct the mislabeled pixels.”
Deep Interactive Thin Object Selection (2021, methodology): “Therefore, at below we ﬁrst introduce a new large-scale dataset for thin object segmen-1 , Latent Diversity [21], f-BRS [41] and DEXTR [32].”
PseudoClick: Interactive Image Segmentation with Click Imitation (2022, methodology): “Binary cross entropy (BCE) loss is widely used to supervise the interactive segmentation tasks [32,33,34,13,35].”
Continuous Adaptation for Interactive Object Segmentation by Learning from Corrections (2019, methodology): “We adopt a typical interactive object segmentation process [12,56,38,33,11,29]: the model is given an image and makes an initial foreground / background prediction for every pixel.”
Interactive Image Segmentation With First Click Attention (2020, methodology): “Following [25, 30, 31, 37, 47], we employ the common FCN architecture, whose specific structure is similar to DeepLab v3+ [7].”