检索结果-内蒙古大学图书馆

Meta-transfer learning through hard tasks

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Sun, Qianru Liu, Yaoyao Chen, Zhaozheng Chua, Tat-Seng Schiele, Bernt School of Information Systems Singapore Management University School of Electrical and Information Engineering Tianjin University School of Computing National University of Singapore Department of Computer Vision and Machine Learning Max-Plank Institute for Informatics

Meta-learning has been proposed as a framework to address the challenging few-shot learning setting. The key idea is to leverage a large number of similar few-shot tasks in order to learn how to adapt a base-learner to a new task for which only a few labeled samples are available. As deep neural networks (DNNs) tend to overfit using a few samples only, typical meta-learning models use shallow neural networks, thus limiting its effectiveness. In order to achieve top performance, some recent works tried to use the DNNs pre-trained on large-scale datasets but mostly in straight-forward manners, e.g., (1) taking their weights as a warm start of meta-training, and (2) freezing their convolutional layers as the feature extractor of base-learners. In this paper, we propose a novel approach called meta-transfer learning (MTL) which learns to transfer the weights of a deep NN for few-shot learning tasks. Specifically, meta refers to training multiple tasks, and transfer is achieved by learning scaling and shifting functions of DNN weights for each task. In addition, we introduce the hard task (HT) meta-batch scheme as an effective learning curriculum that further boosts the learning efficiency of MTL. We conduct few-shot learning experiments and report top performance for five-class few-shot recognition tasks on three challenging benchmarks: miniImageNet, tieredImageNet and Fewshot-CIFAR100 (FC100). Extensive comparisons to related works validate that our MTL approach trained with the proposed HT meta-batch scheme achieves top performance. An ablation study also shows that both components contribute to fast convergence and high accuracy. Compared to the conference version of the paper [1], this version additionally presents (1) the analysis of using different DNN architectures, e.g., ResNet-12, ResNet-18 and ResNet-25;(2) the new MTL variants that adapt our scaling and shifting functions to the classical supervised and the state-of-the-art semi-supervised meta-learners, and

关键词： Deep neural networks

The wildtrack multi-camera person dataset

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Chavdarova, Tatjana Baqué, Pierre Bouquet, Stéphane Maksai, Andrii Jose, Cijo Lettry, Louis Fua, Pascal van Gool, Luc Fleuret, François Machine Learning group Idiap Research Institute École Polytechnique Fédérale de Lausanne CVLab École Polytechnique Fédérale de Lausanne Computer Vision Lab ETH Zurich

People detection methods are highly sensitive to the perpetual occlusions among the targets. As multi-camera set-ups become more frequently encountered, joint exploitation of the across views information would allow for improved detection performances. We provide a large-scale HD dataset named WILDTRACK which finally makes advanced deep learning methods applicable to this problem. The seven-static-camera set-up captures realistic and challenging scenarios of walking people. Notably, its camera calibration with jointly high-precision projection widens the range of algorithms which may make use of this dataset. In aim to help accelerate the research on automatic camera calibration, such annotations also accompany this dataset. Furthermore, the rich-in-appearance visual context of the pedestrian class makes this dataset attractive for monocular pedestrian detection as well, since: the HD cameras are placed relatively close to the people, and the size of the dataset further increases seven-fold. In summary, we overview existing multi-camera datasets and detection methods, enumerate details of our dataset, and we benchmark multi-camera state of the art detectors on this new dataset. Copyright © 2017, The Authors. All rights reserved.

关键词： Deep learning

An Efficient Domain-Incremental learning Approach to Drive in All Weather Conditions

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Mirza, Muhammad Jehanzeb Masana, Marc Possegger, Horst Bischof, Horst Institute of Computer Graphics and Vision Graz University of Technology Austria Christian Doppler Laboratory for Embedded Machine Learning Austria TU Graz SAL Dependable Embedded Systems Lab Silicon Austria Labs Austria

Although deep neural networks enable impressive visual perception performance for autonomous driving, their robustness to varying weather conditions still requires attention. When adapting these models for changed environments, such as different weather conditions, they are prone to forgetting previously learned information. This catastrophic forgetting is typically addressed via incremental learning approaches which usually re-train the model by either keeping a memory bank of training samples or keeping a copy of the entire model or model parameters for each scenario. While these approaches show impressive results, they can be prone to scalability issues and their applicability for autonomous driving in all weather conditions has not been shown. In this paper we propose DISC - Domain Incremental through Statistical Correction - a simple online zero-forgetting approach which can incrementally learn new tasks (i.e. weather conditions) without requiring re-training or expensive memory banks. The only information we store for each task are the statistical parameters as we categorize each domain by the change in first and second order statistics. Thus, as each task arrives, we simply 'plug and play' the statistical vectors for the corresponding task into the model and it immediately starts to perform well on that task. We show the efficacy of our approach by testing it for object detection in a challenging domain-incremental autonomous driving scenario where we encounter different adverse weather conditions, such as heavy rain, fog, and snow. © 2022, CC BY.

关键词： Meteorology

MOLTR: Multiple object localisation, tracking, and reconstruction from monocular RGB videos

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Li, Kejie Rezatofighi, Hamid Reid, Ian School of Computer Science Australian Institute for Machine Learning University of Adelaide Australia Australian Centre for Robotic Vision Australia Department of Data Science and AI Faculty of Information Technology Monash University ClaytonVIC Australia

Semantic aware reconstruction is more advantageous than geometric-only reconstruction for future robotic and AR/VR applications because it represents not only where things are, but also what things are. Object-centric mapping is a task to build an object-level reconstruction where objects are separate and meaningful entities that convey both geometry and semantic information. In this paper, we present MOLTR, a solution to object-centric mapping using only monocular image sequences and camera poses. It is able to localise, track and reconstruct multiple rigid objects in an online fashion when an RGB camera captures a video of the surrounding. Given a new RGB frame, MOLTR firstly applies a monocular 3D detector to localise objects of interest and extract their shape codes representing the object shape in a learnt embedding. Detections are then merged to existing objects in the map after data association. Motion state (i.e. kinematics and the motion status) of each object is tracked by a multiple model Bayesian filter and object shape is progressively refined by fusing multiple shape code. We evaluate localisation, tracking and reconstruction on benchmarking datasets for indoor and outdoor scenes, and show superior performance over previous approaches. Copyright © 2020, The Authors. All rights reserved.

关键词： Mapping

Exploring the potential of collaborative UAV 3D mapping in Kenyan savanna for wildlife research

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Shukla, Vandita Morelli, Luca Trybala, Pawel Remondino, Fabio Gan, Wentian Yu, Yifei Wang, Xin Trento Italy Computer Vision and Machine Learning Systems Group Institute for Geoinformatics University of Muenster Germany Dept. of Civil Environmental and Mechanical Engineering University of Trento Italy School of Geodesy and Geomatics Wuhan University China

UAV-based biodiversity conservation applications have exhibited many data acquisition advantages for researchers. UAV platforms with embedded data processing hardware can support conservation challenges through 3D habitat mapping, surveillance and monitoring solutions. High-quality real-time scene reconstruction as well as real-time UAV localization can optimize the exploration vs exploitation balance of single or collaborative mission. In this work, we explore the potential of two collaborative frameworks - Visual Simultaneous Localization and Mapping (V-SLAM) and Structure-from-Motion (SfM) for 3D mapping purposes and compare results with standard offline approaches. © 2024, CC BY-NC-SA.

关键词： Unmanned aerial vehicles (UAV)

Txt2Img-MHN: Remote Sensing Image Generation from Text Using Modern Hopfield Networks

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Xu, Yonghao Yu, Weikang Ghamisi, Pedram Kopp, Michael Hochreiter, Sepp Vienna1030 Austria Computer Vision Laboratory Department of Electrical Engineering Linköping University Linköping58183 Sweden Helmholtz-Zentrum Dresden-Rossendorf Helmholtz Institute Freiberg for Resource Technology Machine Learning Group Freiberg09599 Germany ELLIS Unit Linz and LIT AI Lab Institute for Machine Learning Johannes Kepler University Linz4040 Austria

The synthesis of high-resolution remote sensing images based on text descriptions has great potential in many practical application scenarios. Although deep neural networks have achieved great success in many important remote sensing tasks, generating realistic remote sensing images from text descriptions is still very difficult. To address this challenge, we propose a novel text-to-image modern Hopfield network (Txt2Img-MHN). The main idea of Txt2Img-MHN is to conduct hierarchical prototype learning on both text and image embeddings with modern Hopfield layers. Instead of directly learning concrete but highly diverse text-image joint feature representations for different semantics, Txt2Img-MHN aims to learn the most representative prototypes from text-image embeddings, achieving a coarse-to-fine learning strategy. These learned prototypes can then be utilized to represent more complex semantics in the text-to-image generation task. To better evaluate the realism and semantic consistency of the generated images, we further conduct zero-shot classification on real remote sensing data using the classification model trained on synthesized images. Despite its simplicity, we find that the overall accuracy in the zero-shot classification may serve as a good metric to evaluate the ability to generate an image from text. Extensive experiments on the benchmark remote sensing text-image dataset demonstrate that the proposed Txt2Img-MHN can generate more realistic remote sensing images than existing methods. Code and pre-trained models are available online (https://***/YonghaoXu/Txt2Img-MHN). Copyright © 2022, The Authors. All rights reserved.

关键词： Remote sensing

Looking Beyond Two Frames: End-to-End Multi-Object Tracking Using Spatial and Temporal Transformers

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Zhu, Tianyu Hiller, Markus Ehsanpour, Mahsa Ma, Rongkai Drummond, Tom Reid, Ian Rezatofighi, Hamid The Department of Electrical and Computer Systems Engineering Monash University Australia The School of Computing and Information Systems The University of Melbourne Australia The Australian Institute for Machine Learning The University of Adelaide Australia The Department of Data Science and AI Monash University Australia The Australian Centre for Robotic Vision Australia

Tracking a time-varying indefinite number of objects in a video sequence over time remains a challenge despite recent advances in the field. Most existing approaches are not able to properly handle multi-object tracking challenges such as occlusion, in part because they ignore long-term temporal information. To address these shortcomings, we present MO3TR: a truly end-to-end Transformer-based online multi-object tracking (MOT) framework that learns to handle occlusions, track initiation and termination without the need for an explicit data association module or any heuristics. MO3TR encodes object interactions into long-term temporal embeddings using a combination of spatial and temporal Transformers, and recursively uses the information jointly with the input data to estimate the states of all tracked objects over time. The spatial attention mechanism enables our framework to learn implicit representations between all the objects and the objects to the measurements, while the temporal attention mechanism focuses on specific parts of past information, allowing our approach to resolve occlusions over multiple frames. Our experiments demonstrate the potential of this new approach, achieving results on par with or better than the current state-of-the-art on multiple MOT metrics for several popular multi-object tracking benchmarks. © 2021, CC BY.

关键词： Tracking (position)

VICE: Variational Interpretable Concept Embeddings

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Muttenthaler, Lukas Zheng, Charles Y. McClure, Patrick Vandermeulen, Robert A. Hebart, Martin N. Pereira, Francisco Machine Learning Group Technische Universität Berlin BIFOLD Berlin Germany Machine Learning Team FMRI Facility National Institute of Mental Health BethesdaMD United States Department of Computer Science Naval Postgraduate School MontereyCA United States Vision and Computational Cognition Group MPI for Human Cognitive and Brain Sciences Leipzig Germany The Max Planck Institute for Human Cognitive and Brain Sciences Leipzig Germany The National Institute of Mental Health BethesdaMD United States

A central goal in the cognitive sciences is the development of numerical models for mental representations of object concepts. This paper introduces Variational Interpretable Concept Embeddings (VICE), an approximate Bayesian method for embedding object concepts in a vector space using data collected from humans in a triplet odd-one-out task. VICE uses variational inference to obtain sparse, non-negative representations of object concepts with uncertainty estimates for the embedding values. These estimates are used to automatically select the dimensions that best explain the data. We derive a PAC learning bound for VICE that can be used to estimate generalization performance or determine a sufficient sample size for experimental design. VICE rivals or outperforms its predecessor, SPoSE, at predicting human behavior in the triplet odd-one-out task. Furthermore, VICE's object representations are more reproducible and consistent across random initializations, highlighting the unique advantage of using VICE for deriving interpretable embeddings from human behavior. © 2022, CC BY-SA.

关键词： Embeddings

Pruning neural network models for gene regulatory dynamics using data and domain knowledge

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Hossain, Intekhab Fischer, Jonas Burkholz, Rebekka Quackenbush, John Department of Biostatistics Harvard T.H. Chan School of Public Health BostonMA02115 United States Dep. for Computer Vision and Machine Learning Max Planck Institute for Informatics Saarbrücken Germany Helmholtz Center CISPA for Information Security Saarbrücken Germany

The practical utility of machine learning models in the sciences often hinges on their interpretability. It is common to assess a model's merit for scientific discovery, and thus novel insights, by how well it aligns with already available domain knowledge-a dimension that is currently largely disregarded in the comparison of neural network models. While pruning can simplify deep neural network architectures and excels in identifying sparse models, as we show in the context of gene regulatory network inference, state-of-the-art techniques struggle with biologically meaningful structure learning. To address this issue, we propose DASH‡, a generalizable framework that guides network pruning by using domain-specific structural information in model fitting and leads to sparser, better interpretable models that are more robust to noise. Using both synthetic data with ground truth information, as well as real-world gene expression data, we show that DASH, using knowledge about gene interaction partners within the putative regulatory network, outperforms general pruning methods by a large margin and yields deeper insights into the biological systems being studied. © 2024, CC BY-NC-SA.

关键词： Neural network models