A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Filters
Dense Gaussian Processes for Few-Shot Segmentation
[article]
2021
arXiv
pre-print
Preserved Fulltext
To this end, we propose a few-shot segmentation method based on dense Gaussian process (GP) regression. ...
Few-shot segmentation is a challenging dense prediction task, which entails segmenting a novel query image given only a small annotated support set. ...
The computations were enabled by resources provided by the Swedish National Infrastructure for Computing (SNIC), partially funded by the Swedish Research Council through grant agreement no. 2018-05973. ...
arXiv:2110.03674v1
fatcat:7updqeutjnfstdipenektmoamq
Doubly Deformable Aggregation of Covariance Matrices for Few-shot Segmentation
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
For the few-shot segmentation task, the main challenge is how to accurately measure the semantic correspondence between the support and query samples with limited training data. ...
Specifically, in this work, we first devise a novel hard example mining mechanism to learn covariance kernels for the Gaussian process. ...
Details of Gaussian Process Gaussian process (GP) is a non-linear and non-parametric Bayesian model for regression and classification [38] . ...
arXiv:2208.00306v1
fatcat:nx4nghitc5bq3pdhwhhm54gfnu
Guided Networks for Few-Shot Image Segmentation and Fully Connected CRFs
2020
Electronics
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Structured output tasks like segmentation are challenging for few-shot learning, due to their being high-dimensional and statistically dependent. ...
The goal of the few-shot learning method is to learn quickly from a low-data regime. ...
In addition, they need to optimize for few-shot usage during the inference process. ...
doi:10.3390/electronics9091508
fatcat:taoh526ygjddpnfwymmzwar7su
Dense Rigid Reconstruction from Unstructured Discontinuous Video
2015
2015 IEEE International Conference on Computer Vision Workshop (ICCVW)
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2015; you can also visit the original URL.
The file type is application/pdf.
The technique is robust against distractors in the scene, background clutter and even shot cuts. ...
We present a novel algorithm for modelling 3D shapes from unstructured, unconstrained discontinuous footage. ...
Coarse Online Modelling For online target/background segmentation, we train a Gaussian Process (GP) as our coarse, probabilistic model. ...
doi:10.1109/iccvw.2015.110
dblp:conf/iccvw/LebedaHB15
fatcat:uoadiaudczckxgvqo7fzk4w3ti
JUMAS @ TRECVID 2010
2010
TREC Video Retrieval Evaluation
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2016; you can also visit the original URL.
The file type is application/pdf.
For Instance Search we submitted a single run where we used the Dutch translation of the queries to retrieve the ASR text. This run reached an AP sum of 0.508 for the 22 topics. ...
We used the the video processing subsystem developed for TRECVID 2009 tasks [1] . We computed a grayscale HOG descriptor on a dense grid for each image. ...
For SIX this formula scores videos. For CLEF features, we obtained prediction for shots and aggregated for videos by averaging after skipping the first and last 5% of the video shots. ...
dblp:conf/trecvid/DaroczyNPPBFG10
fatcat:t7z4m4bs6jealgetkddhn2lgfy
TokyoTech at TRECVID 2015
2015
TREC Video Retrieval Evaluation
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
such as optical-flow value similarity for temporal segmentation. ...
It uses optical-flow features in addition to color or texture features for video segmentation to produce temporally continuous candidate bounding-boxes for object detection. ...
Our future work will focus on deep learning techniques such as deep convolutional neural networks for event detection. ...
dblp:conf/trecvid/InoueTYS15
fatcat:v6yhwbtk5nbalbksyb7dyumn24
Prototype Mixture Models for Few-shot Semantic Segmentation
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Few-shot segmentation is challenging because objects within the support and query images could significantly differ in appearance and pose. ...
Particularly, PMMs improve 5-shot segmentation performance on MS-COCO by up to 5.82\% with only a moderate cost for model size and inference speed. ...
aggregation [1] , with few-shot segmentation. ...
arXiv:2008.03898v2
fatcat:gzejkdc7xbenjpxrhd6oqio6v4
Table of contents
2018
IEEE Transactions on Image Processing
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Ambai 5378 A Unified Approach for Conventional Zero-Shot, Generalized Zero-Shot, and Few-Shot Learning ...................... ........................................................................... ...
Bronstein 5707 Multiresolution Processing of Images and Video Photo Realistic Image Completion via Dense Correspondence ........................... J.-J. Huang and P. L. ...
doi:10.1109/tip.2018.2871548
fatcat:bbsllfph2bcv5lf5wcr4vgedoe
Uncertainty-Aware Semi-Supervised Few Shot Segmentation
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Few shot segmentation (FSS) aims to learn pixel-level classification of a target object in a query image using only a few annotated support samples. ...
We employ the uncertainty estimates to exclude predictions with high degrees of uncertainty for pseudo label construction to obtain additional prototypes based on the refined pseudo labels. ...
Related Works Few Shot Semantic Segmentation Existing few-shot segmentation (FSS) models use the metalearning framework via task-based episodic training of support and query images. ...
arXiv:2110.08954v1
fatcat:idnt7mnmnfh2di73yqpo2juww4
Objectness-Aware Few-Shot Semantic Segmentation
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Few-shot semantic segmentation models aim to segment images after learning from only a few annotated examples. ...
A key challenge for them is how to avoid overfitting because limited training data is available. ...
Few-Shot Semantic Segmentation Results: Results for 1-shot and 5-shot segmentation are shown in Table 1 . ...
arXiv:2004.02945v3
fatcat:7ltlhinjsfd3vks5demzdlqdma
One Shot Learning for Generic Instance Segmentation in RGBD Videos
2019
Proceedings of the 14th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
instance samples via CNNs to generate robust features for instance segmentation. ...
In practice, a classical generic instance segmentation method is employed to initially detect object instances and build temporal correspondences, whereas instance models are trained based on the few detected ...
However, this advantage of CNNs becomes an obstacle in the training process when only few annotation is provided. ...
doi:10.5220/0007259902330239
dblp:conf/visapp/0003CP19
fatcat:gah4s4mhkrbqtc52dhigzje6jy
Contrastive Enhancement Using Latent Prototype for Few-Shot Segmentation
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Few-shot segmentation enables the model to recognize unseen classes with few annotated examples. ...
Extensive experiments show our approach remarkably improves the performance of state-of-the-art methods for 1-shot and 5-shot segmentation, especially outperforming baseline by 5.9% and 7.3% for 5-shot ...
Few-Shot Segmentation Few-Shot Segmentation methods are mainly based on metric learning and extend from fewshot image classification methods. [6] early formulated n-way k-shot few-shot semantic segmentation ...
arXiv:2203.04095v1
fatcat:s7bdq6cshbcz5mgvio2cofw2we
Uncertainty-Aware Learning for Zero-Shot Semantic Segmentation
2020
Neural Information Processing Systems
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Zero-shot semantic segmentation (ZSS) aims to classify pixels of novel classes without training examples available. ...
Specifically, we model the network outputs with Gaussian and Laplacian distributions, with the variances accounting for the observation noise and uncertainty of input samples. ...
Our work aims to learn reliable models for zero-shot semantic segmentation. ...
dblp:conf/nips/HuSS20
fatcat:rn222afzmfdo3fntksbyoky2ry
A Few-Shot Sequential Approach for Object Counting
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
This process is employed on an adapted prototypical-based few-shot approach that uses the extracted features to classify each one either as one of the classes present in the support set images or as background ...
In this work, we address the problem of few-shot multi-class object counting with point-level annotations. ...
Few-shot counting, detection and segmentation: the majority of approaches in few-shot learning focus on the problem of object classification. ...
arXiv:2007.01899v2
fatcat:q52g6jnbujasxandw3mw4zuc5a
PDFNet: Pointwise Dense Flow Network for Urban-Scene Segmentation
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
While regularization techniques come at minimal cost, the collection of labeled data is an expensive and laborious process. ...
The extensive experiments on Cityscapes and CamVid benchmarks demonstrate that our method significantly outperforms baselines in capturing small classes and in few-data regimes. ...
1 The encoder connections in the PDFNet are similar to the full dense connectivity pattern introduced in CondenseNet [64]. ...
arXiv:2109.10083v1
fatcat:4zzppv7iuzelxiqu2i6h5fdiey
« Previous
Showing results 1 — 15 out of 6,885 results