A survey of recent interactive image segmentation methods
Image segmentation is one of the most basic tasks in computer vision and remains an initial
step of many applications. In this paper, we focus on interactive image segmentation (IIS) …
step of many applications. In this paper, we focus on interactive image segmentation (IIS) …
Normalized cut loss for weakly-supervised cnn segmentation
Most recent semantic segmentation methods train deep convolutional neural networks with
fully annotated masks requiring pixel-accuracy for good quality training. Common weakly …
fully annotated masks requiring pixel-accuracy for good quality training. Common weakly …
On regularized losses for weakly-supervised cnn segmentation
Minimization of regularized losses is a principled approach to weak supervision well-
established in deep learning, in general. However, it is largely overlooked in semantic …
established in deep learning, in general. However, it is largely overlooked in semantic …
Lucid data dreaming for video object segmentation
Convolutional networks reach top quality in pixel-level video object segmentation but
require a large amount of training data (1k–100k) to deliver such results. We propose a new …
require a large amount of training data (1k–100k) to deliver such results. We propose a new …
Road extraction from very high resolution images using weakly labeled OpenStreetMap centerline
Road networks play a significant role in modern city management. It is necessary to
continually extract current road structure, as it changes rapidly with the development of the …
continually extract current road structure, as it changes rapidly with the development of the …
Blockchain in IoT security: a survey
Blockchain shows a huge prospective in the coming future. It is atechnology that provides
the possibility of generating and sharing transaction ledgers that are tamper proof. Use …
the possibility of generating and sharing transaction ledgers that are tamper proof. Use …
Kernel Cuts: Kernel and Spectral Clustering Meet Regularization
This work bridges the gap between two popular methodologies for data partitioning: kernel
clustering and regularization-based segmentation. While addressing closely related …
clustering and regularization-based segmentation. While addressing closely related …
Kernel clustering: Density biases and solutions
Kernel methods are popular in clustering due to their generality and discriminating power.
However, we show that many kernel clustering criteria have density biases theoretically …
However, we show that many kernel clustering criteria have density biases theoretically …
Weakly supervised segmentation loss based on graph cuts and superpixel algorithm
M Li, D Chen, S Liu - Neural Processing Letters, 2022 - Springer
In recent years, weakly supervised learning is a hot topic in the field of machine learning,
especially for image segmentation. Assuming that only a small number of pixel categories …
especially for image segmentation. Assuming that only a small number of pixel categories …
Explored normalized cut with random walk refining term for image segmentation
The Normalized Cut (NCut) model is a popular graph-based model for image segmentation.
But it suffers from the excessive normalization problem and weakens the small object and …
But it suffers from the excessive normalization problem and weakens the small object and …