检索结果-内蒙古大学图书馆

CarFormer: Self-Driving with Learned Object-Centric Representations

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Hamdan, Shadi Güney, Fatma Department of Computer Engineering Koç University KUIS AI Center Turkey

The choice of representation plays a key role in self-driving. Bird’s eye view (BEV) representations have shown remarkable performance in recent years. In this paper, we propose to learn object-centric representations in BEV to distill a complex scene into more actionable information for self-driving. We first learn to place objects into slots with a slot attention model on BEV sequences. Based on these object-centric representations, we then train a transformer to learn to drive as well as reason about the future of other vehicles. We found that object-centric slot representations outperform both scene-level and object-level approaches that use the exact attributes of objects. Slot representations naturally incorporate information about objects from their spatial and temporal context such as position, heading, and speed without explicitly providing it. Our model with slots achieves an increased completion rate of the provided routes and, consequently, a higher driving score, with a lower variance across multiple runs, affirming slots as a reliable alternative in object-centric approaches. Additionally, we validate our model’s performance as a world model through forecasting experiments, demonstrating its capability to predict future slot representations accurately. The code and the pre-trained models can be found at https://***/CarFormer/. © 2024, CC BY.

关键词：

IMGTP: A Unified Framework for Improving and Measuring the Generalizability of Trajectory Prediction Models

学校读者我要写书评

暂无评论

IEEE Transactions on Intelligent Vehicles

IEEE Transactions on Intelligent Vehicles 2024年 1-15页

作者： Ye, Luyao Zhou, Zikang Wang, Jianping Li, Yung-Hui Jan, Nien-Yi Lin, Yi-Rong Lin, Yen-Cheng National Engineering Research Center of Educational Big Data and the Faculty of Artificial Intelligence in Education Central China Normal University China Department of Computer Science City University of Hong Kong and City University of Hong Kong Shenzhen Research Institute China AI Research Center of Hon Hai Research Institute China

Accurately predicting nearby agents' future trajectories is fundamental for ensuring the safety and efficiency of autonomous driving. However, existing learning-based trajectory prediction models struggle with poor generalization to out-of-distribution (OOD) scenes, potentially leading to failed predictions in real-world applications. To tackle this challenge, we introduce IMGTP, a unified framework for systematically evaluating, improving, and measuring the generalization performance of trajectory prediction models in OOD scenes. IMGTP consists of three primary components: the Argoverse-Shift dataset, the FrenÃ©t+ strategy, and an approach for OOD measurement. First, the Argoverse-Shift is a cross-domain dataset that highlights the significant differences in data distributions across domains, serving as a testing platform for models' generalization ability. Second, the FrenÃ©t+ strategy leverages map centerlines and the FrenÃ©t Frame to enhance the generalizability of baseline models by mitigating the domain gap resulting from road geometry variations. Third, we provide a comprehensive evaluation methodology for OOD measurement, quantifying the range of data that can be effectively handled by prediction models. This measurement approach employs the accuracy rate of the correctly predicted samples and their corresponding mean distance as well as standard deviation to the training set center as indicators of the model's generalization ability. The mean distance also serves as a valid criterion for OOD detection and uncertainty quantification. Experimental results demonstrate that our IMGTP framework enhances the generalization capability of mainstream trajectory prediction models by substantially improving their prediction performance in unseen domains. This advancement contributes to developing more robust and safer prediction models for autonomous driving in the real world. The code can be found at https://***/XIAOYEJIAYOU/Frenet-Strategy. IEEE

关键词： Autonomous vehicles

Autonomous Point Cloud Segmentation for Power Lines Inspection in Smart Grid

学校读者我要写书评

暂无评论

IFAC-PapersOnLine 2023年第2期56卷 11754-11761页

作者： Alexander Kyuroson Anton Koval George Nikolakopoulos Robotics & AI Team Department of Computer Electrical and Space Engineering Luleå University of Technology Luleå SE-97187 Sweden

LiDAR is currently one of the most utilized sensors to effectively monitor the status of power lines and facilitate the inspection of remote power distribution networks and related infrastructures. To ensure the safe operation of the smart grid, various remote data acquisition strategies, such as airborne Laser Scanning (ALS), Mobile Laser Scanning (MLS), and Terrestrial Laser Scanning (TSL) have been leveraged to allow continuous monitoring of regional power networks, which are typically surrounded by dense vegetation. In this article, an unsupervised Machine Learning (ML) framework is proposed, to detect, extract and analyze the characteristics of power lines of both high and low voltage, as well as the surrounding vegetation in a Power Line Corridor (PLC) solely from LiDAR data. Initially, the proposed approach eliminates the ground points from higher elevation points based on statistical analysis that applies density criteria and histogram thresholding. After denoising and transforming of the remaining candidate points by applying Principle Component Analysis (PCA) and Kd-tree, power line segmentation is achieved by utilizing a two-stage DBSCAN clustering to identify each power line individually. Finally, all high elevation points in the PLC are identified based on their distance to the newly segmented power lines. Conducted experiments illustrate that the proposed framework is an agnostic method that can efficiently detect the power lines and perform PLC-based hazard analysis.

关键词： Power line Segmentation Unsupervised learning DBSCAN UAV

Class-Agnostic Visio-Temporal Scene Sketch Semantic Segmentation

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Kütük, Aleyna Sezgin, Tevfik Metin Department of Computer Engineering KUIS AI Center Koç University Turkey

Scene sketch semantic segmentation is a crucial task for various applications including sketch-to-image retrieval and scene understanding. Existing sketch segmentation methods treat sketches as bitmap images, leading to the loss of temporal order among strokes due to the shift from vector to image format. Moreover, these methods struggle to segment objects from categories absent in the training data. In this paper, we propose a Class-Agnostic Visio-Temporal Network (CAVT) for scene sketch semantic segmentation. CAVT employs a class-agnostic object detector to detect individual objects in a scene and groups the strokes of instances through its post-processing module. This is the first approach that performs segmentation at both the instance and stroke levels within scene sketches. Furthermore, there is a lack of free-hand scene sketch datasets with both instance and stroke-level class annotations. To fill this gap, we collected the largest Free-hand Instance- and Stroke-level Scene Sketch Dataset (FrISS) that contains 1K scene sketches and covers 403 object classes with dense annotations. Extensive experiments on FrISS and other datasets demonstrate the superior performance of our method over state-of-the-art scene sketch segmentation models. The code and dataset will be made public after acceptance. © 2024, CC BY-NC-ND.

关键词： Semantic Segmentation

Cash Flow Forecasting Based on Wavelet Transform and Neural Networks 8

学校读者我要写书评

暂无评论

Cash Flow Forecasting Based on Wavelet Transform and Neural ...

8th International Conference on computer Science and engineering, UBMK 2023

作者： Soylen, Zumra Mammadova, Farida Fasounaki, Mandana Ozmen, Ayse Irem Ince, Gokhan Istanbul Technical University Department of Computer Engineering İstanbul Turkey Mdp Group R&d Department İstanbul Turkey Istanbul Technical University Department of Computer Engineering & Ai and Data Science Application and Research Center İstanbul Turkey

ISBN: (纸本)9798350340815

Cash flow forecasting is a critical task for businesses and financial institutions to ensure effective financial planning and decision-making. However, limited data availability poses a significant challenge when developing accurate and robust cash flow prediction models. In this paper, we investigate the performance of various forecasting methods and propose an approach based on wavelet transform for improving the forecasting accuracy. We demonstrate the effectiveness of the proposed approach with the best combination of wavelet functions and methods for forecasting future values in a univariate time series. We investigate the impact of wavelet transform on forecasting techniques based on open-source datasets. Our methodology includes data collection, preprocessing, feature engineering, model selection, and experimentation using different performance evaluation metrics. © 2023 IEEE.

关键词： Forecasting

O1O: Grouping of Known Classes to Identify Unknown Objects as Odd-One-Out

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Yavuz, Mısra Güney, Fatma Department of Computer Engineering Koç University KUIS AI Center Turkey

Object detection methods trained on a fixed set of known classes struggle to detect objects of unknown classes in the open-world setting. Current fixes involve adding approximate supervision with pseudo-labels corresponding to candidate locations of objects, typically obtained in a class-agnostic manner. While previous approaches mainly rely on the appearance of objects, we find that geometric cues improve unknown recall. Although additional supervision from pseudo-labels helps to detect unknown objects, it also introduces confusion for known classes. We observed a notable decline in the model's performance for detecting known objects in the presence of noisy pseudo-labels. Drawing inspiration from studies on human cognition, we propose to group known classes into superclasses. By identifying similarities between classes within a superclass, we can identify unknown classes through an odd-one-out scoring mechanism. Our experiments on open-world detection benchmarks demonstrate significant improvements in unknown recall, consistently across all tasks. Crucially, we achieve this without compromising known performance, thanks to better partitioning of the feature space with superclasses. Project page: https://***/O1O. © 2024, CC BY.

关键词： Object detection

Neurocache: Efficient Vector Retrieval for Long-range Language Modeling

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Safaya, Ali Yuret, Deniz KUIS AI Center Computer Engineering Department Koç University Turkey

This paper introduces Neurocache, an approach to extend the effective context size of large language models (LLMs) using an external vector cache to store its past states. Like recent vector retrieval approaches, Neurocache uses an efficient k-nearest-neighbor (kNN) algorithm to retrieve relevant past states and incorporate them into the attention process. Neurocache improves upon previous methods by (1) storing compressed states, which reduces cache size;(2) performing a single retrieval operation per token which increases inference speed;and (3) extending the retrieval window to neighboring states, which improves both language modeling and downstream task accuracy. Our experiments show the effectiveness of Neurocache both for models trained from scratch and for pre-trained models such as Llama2-7B and Mistral-7B when enhanced with the cache mechanism. We also compare Neurocache with text retrieval methods and show improvements in single-document question-answering and few-shot learning tasks. We made the source code available under: https://***/alisafaya/neurocache © 2024, CC BY.

关键词： Nearest neighbor search

DRONECOM: Dynamic Relay Operations for Network Efficient Communication in Mines

学校读者我要写书评

暂无评论

DRONECOM: Dynamic Relay Operations for Network Efficient Com...

International Conference on Control, Automation and Systems ( ICCAS)

作者： Akash Patel Scott Fredriksson Samuel Nordström Emanuele Pagliari Ilias Tevetzidis Christoforos Kanellakis George Nikolakopoulos Department of Computer Science Electrical and Space Engineering Robotics & AI Team Luleå University of Technology Luleå Sweden

ISBN: (数字)9788993215380

ISBN: (纸本)9798331517939

As the routine operations are starting to become highly automated, it is crucial to develop autonomous solutions that are infrastructure independent. Achieving this is challenging due to the ever-changing landscape of mines, which complicates infrastructure development. In response, this paper introduces a robust framework employing drones to gather data from hard-to-access areas in mines and deliver the data back to the base station for routine monitoring purposes. These tasks include gathering data from operational vehicles (mine trucks, loaders etc.), as well as various sensors (e.g. monitoring rock bolts) and relaying the data to the mine’s base station for monitoring purposes. The proposed framework is based on autonomous navigation using a known point cloud map of the mine, proximity detection via Ultra WideBand (UWB) radios and the data transfer is accomplished through the IEEE 802.15.4 communication standard, operating in the 868 MHz ISM band, with the aim to guarantee long range operation. On the mission level, the drones act as data mules capable of autonomously extracting data from operating vehicles, storing the data onboard and eventually delivering the data to the base station, which is enabled through a Point and Click (PAC) autonomy framework based on global planning, reactive navigation, communication link and behavior management. The efficacy of this framework has been demonstrated through real-world experiments conducted at a test mine in Sweden, validating the overall architecture of the proposed solution.

关键词： Base stations Automation Navigation Fasteners Rocks Sensors Data mining Monitoring Drones Videos

Towards efficient generative ai and beyond-ai computing:New trends on ISSCC 2024 machine learning accelerators

学校读者我要写书评

暂无评论

Journal of Semiconductors 2024年第4期45卷 12-15页

作者： Bohan Yang Jia Chen Fengbin Tu Department of Electronic and Computer Engineering The Hong Kong University of Science and TechnologyHong KongChina AI Chip Center for Emerging Smart Systems The Hong Kong University of Science and TechnologyHong KongChina School of the Gifted Young University of Science and Technology of ChinaHefei 230026Chin

Compared to the last decade when the convolution neu-ral network(CNN)dominated the research field,machine learn-ing(ML)algorithms have reached a pivotal moment called the generative artificial intelligence(ai)*** the emer-gence of large-scale foundation models[1],such as large multi-modal model(LMM)GPT-4[2]and text-to-image generative model DALL·E[3].

关键词： ISSCC beyond ai