Filters








8,699 Hits in 5.4 sec

Revisiting Few-shot Activity Detection with Class Similarity Control [article]

Huijuan Xu, Ximeng Sun, Eric Tzeng, Abir Das, Kate Saenko, Trevor Darrell
2020 arXiv   pre-print
In this paper, we present a conceptually simple and general yet novel framework for few-shot temporal activity detection based on proposal regression which detects the start and end time of the activities  ...  We experiment on three large scale benchmarks for temporal activity detection (ActivityNet1.2, ActivityNet1.3 and THUMOS14 datasets) in a few-shot setting.  ...  Additionally, we also show the effect of controlling the novel class similarity with the pre-training classes on few-shot detection performance. (1) a 3D ConvNet feature extractor applied to both the untrimmed  ... 
arXiv:2004.00137v1 fatcat:fi4cbvma35bblhy6dzhaab3py4

Revisiting Few-Shot Learning for Facial Expression Recognition [article]

Anca-Nicoleta Ciubotaru, Arnout Devos, Behzad Bozorgtabar, Jean-Philippe Thiran, Maria Gabrani
2019 arXiv   pre-print
In this paper, we revisit and compare existing few-shot learning methods for the low-shot facial expression recognition in terms of their generalisation ability via episode-training.  ...  We embrace these challenges and formulate the problem as a low-shot learning, where once the base classifier is deployed, it must rapidly adapt to recognise novel classes using a few samples.  ...  In Table I we saw that few-shot learning algorithms are able to generalise better when the novel classes are extracted from a dataset created in a controlled environment, thus with less intra-class variation  ... 
arXiv:1912.02751v2 fatcat:nfaifcpoxjcozhzrgxzd24mbmi

Incremental Few-Shot Object Detection [article]

Juan-Manuel Perez-Rua and Xiatian Zhu and Timothy Hospedales and Tao Xiang
2020 arXiv   pre-print
We present a study aiming to go beyond these limitations by considering the Incremental Few-Shot Detection (iFSD) problem setting, where new classes must be registered incrementally (without revisiting  ...  base classes) and with few examples.  ...  For few-shot object detection in particular, a key merit of CentreNet is that each individual class maintains its own prediction heatmap and makes independent detection by activation thresholding.  ... 
arXiv:2003.04668v2 fatcat:kzmyl25s4vdxzbwmgh4cvcwufe

Incremental Few-Shot Object Detection

Juan-Manuel Perez-Rua, Xiatian Zhu, Timothy M. Hospedales, Tao Xiang
2020 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
We present a study aiming to go beyond these limitations by considering the Incremental Few-Shot Detection (iFSD) problem setting, where new classes must be registered incrementally (without revisiting  ...  base classes) and with few examples.  ...  For few-shot object detection in particular, a key merit of CentreNet is that each individual class maintains its own prediction heatmap and makes independent detection by activation thresholding.  ... 
doi:10.1109/cvpr42600.2020.01386 dblp:conf/cvpr/Perez-RuaZHX20 fatcat:76xg2hmjkrg67kro3e6qfd6t3y

Wandering Within a World: Online Contextualized Few-Shot Learning [article]

Mengye Ren, Michael L. Iuzzolino, Michael C. Mozer, Richard S. Zemel
2021 arXiv   pre-print
We aim to bridge the gap between typical human and machine-learning environments by extending the standard framework of few-shot learning to an online, continual setting.  ...  Building upon this setting, we propose a new few-shot learning dataset based on large scale indoor imagery that mimics the visual experience of an agent wandering within a world.  ...  To break the rigid, artificial structure of continual and few-shot learning, we propose a new continual few-shot learning setting where environments are revisited and the total number of novel object classes  ... 
arXiv:2007.04546v3 fatcat:5giqdg6gbbdqvj4ekzizlo7rkm

DeFRCN: Decoupled Faster R-CNN for Few-Shot Object Detection [article]

Limeng Qiao, Yuxuan Zhao, Zhiyuan Li, Xi Qiu, Jianan Wu, Chi Zhang
2021 arXiv   pre-print
Few-shot object detection, which aims at detecting novel objects rapidly from extremely few annotated examples of previously unseen classes, has attracted significant research interest in the community  ...  Extensive experiments on multiple benchmarks show our framework is remarkably superior to other existing approaches and establishes a new state-of-the-art in few-shot literature.  ...  Generalized Few-Shot Object Detection A. 1 .  ... 
arXiv:2108.09017v1 fatcat:w3ntxyh3wbdq7bwrtovgy5x47u

Long-Tailed Classification by Keeping the Good and Removing the Bad Momentum Causal Effect [article]

Kaihua Tang, Jianqiang Huang, Hanwang Zhang
2021 arXiv   pre-print
As the class size grows, maintaining a balanced dataset across many classes is challenging because the data are long-tailed in nature; it is even impossible when the sample-of-interest co-exists with each  ...  The test and val sets were balanced and reported on four splits: Many-shot containing classes with > 100 images, Medium-shot including classes with ≥ 20 & ≤ 100 images, Few-shot covering classes with <  ...  In train set, the number of images per class is ranged from 1,280 to 5, which Many-shot Classes house finch brown bear meerkat alligator lizard Medium-shot Classes Few-shot Classes warthog bighorn  ... 
arXiv:2009.12991v4 fatcat:iirc3ar72bhelm7kjv2thcssxa

Tip-Adapter: Training-free CLIP-Adapter for Better Vision-Language Modeling [article]

Renrui Zhang, Rongyao Fang, Wei Zhang, Peng Gao, Kunchang Li, Jifeng Dai, Yu Qiao, Hongsheng Li
2021 arXiv   pre-print
To further enhance CLIP's few-shot capability, CLIP-Adapter proposed to fine-tune a lightweight residual feature adapter and significantly improves the performance for few-shot classification.  ...  Moreover, the performance of Tip-Adapter can be further boosted by fine-tuning such properly initialized adapter for only a few epochs with super-fast convergence speed.  ...  In Eq. ( 5 ), β in the activation function ϕ controls the sharpness of the affinities.  ... 
arXiv:2111.03930v2 fatcat:ntojz5cn65eghfqvbmcgij4s2i

Putting a bug in ML: The moth olfactory network learns to read MNIST [article]

Charles B. Delahunt, J. Nathan Kutz
2019 arXiv   pre-print
We show that MothNet successfully learns to read given very few training samples (1 to 10 samples per class).  ...  In this few-samples regime, it outperforms standard machine learning methods such as nearest-neighbors, support-vector machines, and neural networks (NNs), and matches specialized one-shot transfer-learning  ...  N refers to the number of training samples per class.  ... 
arXiv:1802.05405v3 fatcat:iuiwuadzjbbs5e7puwkijicagq

Associative Alignment for Few-shot Image Classification [article]

Arman Afrasiyabi, Jean-François Lalonde, Christian Gagné
2020 arXiv   pre-print
Few-shot image classification aims at training a model from only a few examples for each of the "novel" classes.  ...  Experiments on four standard datasets and three backbones demonstrate that combining our centroid-based alignment loss results in absolute accuracy improvements of 4.4%, 1.2%, and 6.2% in 5-shot learning  ...  .: Revisiting local descriptor based image-to-class measure for few-shot learning. In: The Conference on Computer Vision and Pattern Recognition (2019) 26.  ... 
arXiv:1912.05094v3 fatcat:o7lxo6kvijbw3kjb6ekf6qytoa

Overcoming Classifier Imbalance for Long-tail Object Detection with Balanced Group Softmax [article]

Yu Li, Tao Wang, Bingyi Kang, Sheng Tang, Chunfeng Wang, Jintao Li, Jiashi Feng
2020 arXiv   pre-print
We find existing detection methods are unable to model few-shot classes when the dataset is extremely skewed, which can result in classifier imbalance in terms of parameter magnitude.  ...  Solving long-tail large vocabulary object detection with deep learning based models is a challenging and demanding task, which is however under-explored.In this work, we provide the first systematic analysis  ...  are severely imbalanced, since low-shot categories get few chances to be activated.  ... 
arXiv:2006.10408v1 fatcat:6kvftja62zg7bkaqdkgrb4hai4

Partial Is Better Than All: Revisiting Fine-tuning Strategy for Few-shot Learning [article]

Zhiqiang Shen and Zechun Liu and Jie Qin and Marios Savvides and Kwang-Ting Cheng
2021 arXiv   pre-print
The goal of few-shot learning is to learn a classifier that can recognize unseen classes from limited support data with labels.  ...  to novel data, i.e. learning to transfer in few-shot scenario.) or meta-learning.  ...  Considering an N -way K-shot few-shot task, where the support set on novel class has N classes with K labeled images and the query set contains the same N classes with Q unlabeled images in each class,  ... 
arXiv:2102.03983v1 fatcat:u37noahnpze4jm6ba4pcqihdyu

Single-shot carrier–envelope phase measurement of few-cycle laser pulses

T. Wittmann, B. Horvath, W. Helml, M. G. Schätzel, X. Gu, A. L. Cavalieri, G. G. Paulus, R. Kienberger
2009 Nature Physics  
Here, we demonstrate the first single-shot CEP measurement of intense few-cycle laser pulses.  ...  We focus a laser pulse on a gas target and detect photoelectrons emitted in opposing directions ('left-right') parallel to the polarization of the laser. By comparing the left-right asymmetries  ...  Typically, the rate of change of the CEP is detected with a so-called f-to-2f interferometer 2,3,17 , and the CEP drift is then readjusted with a control loop.  ... 
doi:10.1038/nphys1250 fatcat:35gpcetuw5ajteym6ahti2ad7q

A Comprehensive Survey of Few-shot Learning: Evolution, Applications, Challenges, and Opportunities [article]

Yisheng Song, Ting Wang, Subrota K Mondal, Jyoti Prakash Sahoo
2022 arXiv   pre-print
For the sake of avoiding conceptual confusion, we first elaborate and compare a set of similar concepts including few-shot learning, transfer learning, and meta-learning.  ...  Few-shot learning (FSL) has emerged as an effective learning method and shows great potential.  ...  Few-shot Object Detection Few-Shot Object Detection (FSOD) is the task of detecting rare objects from several samples.  ... 
arXiv:2205.06743v1 fatcat:ouqpe2swrbch5n2l5jvjs2y6dm

Few-Shot Object Detection: A Survey [article]

Mona Köhler, Markus Eisenbach, Horst-Michael Gross
2021 arXiv   pre-print
To avoid the need to acquire and annotate these huge amounts of data, few-shot object detection aims to learn from few object instances of new categories in the target domain.  ...  As a result, we emphasize common challenges in evaluation and identify the most promising current trends in this emerging field of few-shot object detection.  ...  In the following, we will briefly discuss differences and similarities with few-shot object de- Zero-shot object detection can be defined similar to few-shot tection.  ... 
arXiv:2112.11699v1 fatcat:d6iubz4ui5abvdccjlkm667vay
« Previous Showing results 1 — 15 out of 8,699 results