检索结果-内蒙古大学图书馆

Torchbearer: A model fitting library for PyTorch

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Harris, Ethan Painter, Matthew Hare, Jonathon Vision Learning and Control Group Department of Electronics and Computer Science University of Southampton United Kingdom

We introduce torchbearer, a model fitting library for pytorch aimed at researchers working on deep learning or differentiable programming. The torchbearer library provides a high level metric and callback API that can be used for a wide range of applications. We also include a series of built in callbacks that can be used for: Model persistence, learning rate decay, logging, data visualization and more. The extensive documentation includes an example library for deep learning and dynamic programming problems and can be found at http://***. The code is licensed under the MIT License and available at https://***/ecs-vlc/torchbearer. Copyright © 2018, The Authors. All rights reserved.

关键词： Open source software

Depth Estimation for a Single Omnidirectional Image with Reversed-Gradient Warming-up Thresholds Discriminator

学校读者我要写书评

暂无评论

Depth Estimation for a Single Omnidirectional Image with Rev...

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Yihong Wu Yuwen Heng Mahesan Niranjan Hansung Kim Vision Learning and Control Research Group School of Electronics and Computer Science University of Southampton UK

Depth estimation for single image using deep learning requires a large labelled depth dataset with various scenes for training. However, currently published omnidirectional depth datasets cover limited types of scenes and are not suitable for depth estimation for various real-world scenes. With the challenge of labelled real-world datasets generation and stability of the performance, we propose an architecture with the Reverse-gradient Warming-up Threshold Discriminator (RWTD) to estimate real-world depth maps from the synthetic ground truth. It takes labelled synthetic scenes of a source domain and unlabelled real-world scenes of a target domain as inputs to predict the corresponding depth maps. Compared with state-of-the-art encoder-decoder models, the proposed architecture shows an 11% points improvement on the testing dataset for depth accuracy.

关键词： Training Deep learning Estimation Signal processing Predictive models Stability analysis Reliability

Mask3D: Mask Transformer for 3D Semantic Instance Segmentation

学校读者我要写书评

暂无评论

Mask3D: Mask Transformer for 3D Semantic Instance Segmentati...

IEEE International Conference on Robotics and Automation (ICRA)

作者： Jonas Schult Francis Engelmann Alexander Hermans Or Litany Siyu Tang Bastian Leibe Computer Vision Group RWTH Aachen University Germany Computer Vision and Learning Group ETH Zürich Switzerland ETH AI Center Zürich Switzerland NVIDIA Santa Clara USA

Modern 3D semantic instance segmentation approaches predominantly rely on specialized voting mechanisms followed by carefully designed geometric clustering techniques. Building on the successes of recent Transformer-based methods for object detection and image segmentation, we propose the first Transformer-based approach for 3D semantic instance segmentation. We show that we can leverage generic Transformer building blocks to directly predict instance masks from 3D point clouds. In our model - called Mask3D - each object instance is represented as an instance query. Using Transformer decoders, the instance queries are learned by iteratively attending to point cloud features at multiple scales. Combined with point features, the instance queries directly yield all instance masks in parallel. Mask3D has several advantages over current state-of-the-art approaches, since it neither relies on (1) voting schemes which require hand-selected geometric properties (such as centers) nor (2) geometric grouping mechanisms requiring manually-tuned hyper-parameters (e.g. radii) and (3) enables a loss that directly optimizes instance masks. Mask3D sets a new state-of-the-art on ScanNet test (+6.2mAP), S3DIS 6-fold (+10.1 mAP), STPLS3D (+11.2 mAP) and ScanNet200 test (+12.4 mAP).

关键词：

LETHA: learning from high quality inputs for 3D pose estimation in low quality images 2

学校读者我要写书评

暂无评论

LETHA: Learning from high quality inputs for 3D pose estimat...

2014 2nd International Conference on 3D vision, 3DV 2014

作者： Penate-Sanchez, Adrianrr Moreno-Noguer, Francesc Andrade-Cetto, Juan Fleuret, François Institut de Robòtica i Informàtica Industrial CSIC-UPC Barcelona Spain Computer Vision and Learning Group Idiap Research Institute Martigny Switzerland École Polytechnique Fédérale de Lausanne Lausanne Switzerland

ISBN: (纸本)9781479970018

We introduce LETHA (learning on Easy data, Test on Hard), a new learning paradigm consisting of building strong priors from high quality training data, and combining them with discriminative machine learning to deal with lowquality test data. Our main contribution is an implementation of that concept for pose estimation. We first automatically build a 3D model of the object of interest from high-definition images, and devise from it a pose-indexed feature extraction scheme. We then train a single classifier to process these feature vectors. Given a low quality test image, we visit many hypothetical poses, extract features consistently and evaluate the response of the classifier. Since this process uses locations recorded during learning, it does not require matching points anymore. We use a boosting procedure to train this classifier common to all poses, which is able to deal with missing features, due in this context to self-occlusion. Our results demonstrate that the method combines the strengths of global image representations, discriminative even for very tiny images, and the robustness to occlusions of approaches based on local feature point descriptors. © 2014 IEEE.

关键词： Image representation

SliceFormer: Deep Dense Depth Estimation from a Single Indoor Omnidirectional Image Using a Slice-Based Transformer

学校读者我要写书评

暂无评论

SliceFormer: Deep Dense Depth Estimation from a Single Indoo...

International Conference on Electronics, Information and Communications (ICEIC)

作者： Yihong Wu Yuwen Heng Mahesan Niranjan Hansung Kim Vision Learning and Control Research Group School of Electronics and Computer Science University of Southampton Southampton United Kingdom

In this research, we tackle the task of estimating depth from a single indoor omnidirectional image. Acknowledging gravity's critical influence in artificially constructed indoor environments, we process the input from equirectangular projection by dividing it into vertical slices. These slices are then utilized as patch embeddings for the transformer encoder, a strategy designed to recreate an equirectangular depth map. Our architecture is evaluated against leading models using real-world datasets, namely Stanford2D3D and Matterport3D, demonstrating its superior performance. These results underscore the significance of our gravity-aligned approach for depth estimation in omnidirectional images, especially in man-made settings.

关键词：

RayNet: learning volumetric 3d reconstruction with ray potentials

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Paschalidou, Despoina Ulusoy, Ali Osman Schmitt, Carolin Van Gool, Luc Geiger, Andreas Autonomous Vision Group MPI for Intelligent Systems Tübingen Microsoft Computer Vision Lab ETH Zürich KU Leuven Computer Vision and Geometry Group ETH Zürich Max Planck ETH Center for Learning Systems

In this paper, we consider the problem of reconstructing a dense 3D model using images captured from different views. Recent methods based on convolutional neural networks (CNN) allow learning the entire task from data. However, they do not incorporate the physics of image formation such as perspective geometry and occlusion. Instead, classical approaches based on Markov Random Fields (MRF) with ray-potentials explicitly model these physical processes, but they cannot cope with large surface appearance variations across different viewpoints. In this paper, we propose RayNet, which combines the strengths of both frameworks. RayNet integrates a CNN that learns view-invariant feature representations with an MRF that explicitly encodes the physics of perspective projection and occlusion. We train RayNet end-to-end using empirical risk minimization. We thoroughly evaluate our approach on challenging real-world datasets and demonstrate its benefits over a piece-wise trained baseline, hand-crafted models as well as other learning-based approaches. Copyright © 2019, The Authors. All rights reserved.

关键词： Image reconstruction

Multi-Vehicle Trajectory Prediction at Intersections using State and Intention Information

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Zhu, Dekai Khan, Qadeer Cremers, Daniel Computer Vision Group CIT Technical University of Munich This work was funded by the Munich Center for Machine Learning Germany

Traditional approaches to prediction of future trajectory of road agents rely on knowing information about their past trajectory. This work rather relies only on having knowledge of the current state and intended direction to make predictions for multiple vehicles at intersections. Furthermore, message passing of this information between the vehicles provides each one of them a more holistic overview of the environment allowing for a more informed prediction. This is done by training a neural network which takes the state and intent of the multiple vehicles to predict their future trajectory. Using the intention as an input allows our approach to be extended to additionally control the multiple vehicles to drive towards desired paths. Experimental results demonstrate the robustness of our approach both in terms of trajectory prediction and vehicle control at intersections. The complete training and evaluation code for this work is available here: https:// ***/Dekai21/Multi_Agent_Intersection. Copyright © 2023, The Authors. All rights reserved.

关键词： Forecasting

Mask3D: Mask Transformer for 3D Semantic Instance Segmentation

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Schult, Jonas Engelmann, Francis Hermans, Alexander Litany, Or Tang, Siyu Leibe, Bastian Computer Vision Group RWTH Aachen University Germany Computer Vision and Learning Group ETH Zürich Switzerland ETH AI Center Zürich Switzerland NVIDIA Santa Clara United States

Modern 3D semantic instance segmentation approaches predominantly rely on specialized voting mechanisms followed by carefully designed geometric clustering techniques. Building on the successes of recent Transformer-based methods for object detection and image segmentation, we propose the first Transformer-based approach for 3D semantic instance segmentation. We show that we can leverage generic Transformer building blocks to directly predict instance masks from 3D point clouds. In our model – called Mask3D – each object instance is represented as an instance query. Using Transformer decoders, the instance queries are learned by iteratively attending to point cloud features at multiple scales. Combined with point features, the instance queries directly yield all instance masks in parallel. Mask3D has several advantages over current state-of-the-art approaches, since it neither relies on (1) voting schemes which require hand-selected geometric properties (such as centers) nor (2) geometric grouping mechanisms requiring manually-tuned hyper-parameters (e.g. radii) and (3) enables a loss that directly optimizes instance masks. Mask3D sets a new state-of-the-art on ScanNet test (+ 6.2 mAP), S3DIS 6-fold (+ 10.1 mAP), STPLS3D (+ 11.2 mAP) and ScanNet200 test (+ 12.4 mAP). © 2022, CC BY.

关键词： Semantics

Spatially Perturbed Collision Sounds Attenuate Perceived Causality in 3D Launching Events

学校读者我要写书评

暂无评论

Spatially Perturbed Collision Sounds Attenuate Perceived Cau...

IEEE Annual International Symposium Virtual Reality

作者： Duotun Wang James Kubrlcht Yixin Zhu Wei Lianq Song-Chun Zhu Chenfanfu Jiang Hongjing Lu Laboratory of Intelligent Information Technology Beijing Institute of Technology UCLA Computational Vision and Learning Laboratory Cognition Center for Vision Computer Graphics Group UPenn

ISBN: (纸本)9781538633663

When a moving object collides with an object at rest, people immediately perceive a causal event: i.e., the first object has launched the second object forwards. However, when the second object's motion is delayed, or is accompanied by a collision sound, causal impressions attenuate and strengthen. Despite a rich literature on causal perception, researchers have exclusively utilized 2D visual displays to examine the launching effect. It remains unclear whether people are equally sensitive to the spatiotemporal properties of observed collisions in the real world. The present study first examined whether previous findings in causal perception with audiovisual inputs can be extended to immersive 3D virtual environments. We then investigated whether perceived causality is influenced by variations in the spatial position of an auditory collision indicator. We found that people are able to localize sound positions based on auditory inputs in VR environments, and spatial discrepancy between the estimated position of the collision sound and the visually observed impact location attenuates perceived causality.

关键词： Three-dimensional displays Visualization Solid modeling Delays Spatiotemporal phenomena Virtual environments