检索结果-内蒙古大学图书馆

Training β-VAE by Aggregating a Learned Gaussian Posterior with a Decoupled Decoder

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Li, Jianning Fragemann, Jana Ahmadi, Seyed-Ahmad Kleesiek, Jens Egger, Jan Institute for AI in Medicine University Medicine Essen Essen Germany Institute of Computer Graphics and Vision Graz University of Technology Graz Austria NVIDIA Munich Germany

The reconstruction loss and the Kullback-Leibler divergence (KLD) loss in a variational autoencoder (VAE) often play antagonistic roles, and tuning the weight of the KLD loss in β-VAE to achieve a balance between the two losses is a tricky and dataset-specific task. As a result, current practices in VAE training often result in a trade-off between the reconstruction fidelity and the continuity/disentanglement of the latent space, if the weight β is not carefully tuned. In this paper, we present intuitions and a careful analysis of the antagonistic mechanism of the two losses, and propose, based on the insights, a simple yet effective two-stage method for training a VAE. Specifically, the method aggregates a learned Gaussian posterior z ∼ qθ(z|x) with a decoder decoupled from the KLD loss, which is trained to learn a new conditional distribution pφ(x|z) of the input data x. Experimentally, we show that the aggregated VAE maximally satisfies the Gaussian assumption about the latent space, while still achieves a reconstruction error comparable to when the latent space is only loosely regularized by N(0, I). The proposed approach does not require hyperparameter (i.e., the KLD weight β) tuning given a specific dataset as required in common VAE training practices. We evaluate the method using a medical dataset intended for 3D skull reconstruction and shape completion, and the results indicate promising generative capabilities of the VAE trained using the proposed method. Besides, through guided manipulation of the latent variables, we establish a connection between existing autoencoder (AE)-based approaches and generative approaches, such as VAE, for the shape completion problem. Codes and pre-trained weights are available at https://***/Jianningli/skullVAE. © 2022, CC BY-NC-SA.

关键词： Economic and social effects

Enabling and Assessing Trust when Cooperating with Robots in Disaster Response (EASIER)

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Frering, Laurent Eder, Matthias Kubicek, Bettina Albert, Dietrich Kalkofen, Denis Gschwandtner, Thomas Krajnz, Heimo Steinbauer-Wagner, Gerald The Institute of Software Technology Graz University of Technology Graz Austria The Institute of Psychology University of Graz Graz Austria The Institute of Computer Graphics and Vision Graz University of Technology Graz Austria Rosenbauer International AG Linz Austria Professional Fire Brigade Graz Graz Austria

This paper presents a conceptual overview of the EASIER project and its scope. EASIER focuses on supporting emergency forces in disaster response scenarios with a semi-autonomous mobile manipulator. Specifically, we examine the operator's trust in the system and his/her cognitive load generated by its use. We plan to address different research topics, exploring how shared autonomy, interaction design, and transparency relate to trust and cognitive load. Another goal is to develop non-invasive methods to continuously measure trust and cognitive load in the context of disaster response using a multilevel approach. This project is conducted by multiple academic partners specializing in artificial intelligence, interaction design, and psychology, as well as an industrial partner for disaster response equipment and end-users for framing the project and the experiments in real use-cases. © 2022, CC BY.

关键词： Noninvasive medical procedures

IB-MVS: An Iterative Algorithm for Deep Multi-View Stereo based on Binary Decisions 32

学校读者我要写书评

暂无评论

IB-MVS: An Iterative Algorithm for Deep Multi-View Stereo ba...

32nd British Machine vision Conference, BMVC 2021

作者： Sormann, Christian Rossi, Mattia Kuhn, Andreas Fraundorfer, Friedrich Institute of Computer Graphics and Vision Graz University of Technology Austria Sony Europe B.V. R&D Center Stuttgart Laboratory 1 Germany

We present a novel deep-learning-based method for Multi-View Stereo. Our method estimates high resolution and highly precise depth maps iteratively, by traversing the continuous space of feasible depth values at each pixel in a binary decision fashion. The decision process leverages a deep-network architecture: this computes a pixelwise binary mask that establishes whether each pixel actual depth is in front or behind its current iteration individual depth hypothesis. Moreover, in order to handle occluded regions, at each iteration the results from different source images are fused using pixelwise weights estimated by a second network. Thanks to the adopted binary decision strategy, which permits an efficient exploration of the depth space, our method can handle high resolution images without trading resolution and precision. This sets it apart from most alternative learning-based Multi-View Stereo methods, where the explicit discretization of the depth space requires the processing of large cost volumes. We compare our method with state-of-the-art Multi-View Stereo methods on the DTU, Tanks and Temples and the challenging ETH3D benchmarks and show competitive results. © 2021. The copyright of this document resides with its authors.

关键词： Pixels

FC-DCNN: A densely connected neural network for stereo estimation

学校读者我要写书评

暂无评论

FC-DCNN: A densely connected neural network for stereo estim...

International Conference on Pattern Recognition

作者： Dominik Hirner Friedrich Fraundorfer Institute for Computer Graphics and Vision Graz University of Technology Austria

We propose a novel lightweight network for stereo estimation. Our network consists of a fully-convolutional densely connected neural network (FC-DCNN) that computes matching costs between rectified image pairs. Our FC-DCNN method learns expressive features and performs some simple but effective postprocessing steps. The densely connected layer structure connects the output of each layer to the input of each subsequent layer. This network structure and the fact that we do not use any fully-connected layers or 3D convolutions leads to a very lightweight network. The output of this network is used in order to calculate matching costs and create a cost-volume. Instead of using time and memory-inefficient cost-aggregation methods such as semiglobal matching or conditional random fields in order to improve the result, we rely on filtering techniques, namely median filter and guided filter. By computing a left-right consistency check we get rid of inconsistent values. Afterwards we use a watershed foreground-background segmentation on the disparity image with removed inconsistencies. This mask is then used to refine the final prediction. We show that our method works well for both challenging indoor and outdoor scenes by evaluating it on the Middlebury, KITTI and ETH3D benchmarks respectively. Our full framework is available at https://***/thedodo/FC-DCNN.

关键词： Matched filters Image segmentation Three-dimensional displays Filtering Neural networks Estimation Benchmark testing

Sit Back and Relax: Learning to Drive Incrementally in All Weather Conditions

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Leitner, Stefan Mirza, M. Jehanzeb Lin, Wei Micorek, Jakub Masana, Marc Kozinski, Mateusz Possegger, Horst Bischof, Horst Institute for Computer Graphics and Vision Graz University of Technology Austria Christian Doppler Laboratory for Embedded Machine Learning Austria Christian Doppler Laboratory for Semantic 3D Computer Vision Austria TU Graz SAL Dependable Embedded Systems Lab Silicon Austria Labs Austria

In autonomous driving scenarios, current object detection models show strong performance when tested in clear weather. However, their performance deteriorates significantly when tested in degrading weather conditions. In addition, even when adapted to perform robustly in a sequence of different weather conditions, they are often unable to perform well in all of them and suffer from catastrophic forgetting. To efficiently mitigate forgetting, we propose Domain-Incremental Learning through Activation Matching (DILAM), which employs unsupervised feature alignment to adapt only the affine parameters of a clear weather pre-trained network to different weather conditions. We propose to store these affine parameters as a memory bank for each weather condition and plug-in their weather-specific parameters during driving (i.e. test time) when the respective weather conditions are encountered. Our memory bank is extremely lightweight, since affine parameters account for less than 2% of a typical object detector. Furthermore, contrary to previous domain-incremental learning approaches, we do not require the weather label when testing and propose to automatically infer the weather condition by a majority voting linear classifier. © 2023, CC BY.

关键词： Object detection

InfoSeg: Unsupervised semantic image segmentation with mutual information maximization

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Harb, Robert Knöbelreiter, Patrick Institute of Computer Graphics and Vision Graz University of Technology Austria

We propose a novel method for unsupervised semantic image segmentation based on mutual information maximization between local and global high-level image features. The core idea of our work is to leverage recent progress in self-supervised image representation learning. Representation learning methods compute a single high-level feature capturing an entire image. In contrast, we compute multiple high-level features, each capturing image segments of one particular semantic class. To this end, we propose a novel two-step learning procedure comprising a segmentation and a mutual information maximization step. In the first step, we segment images based on local and global features. In the second step, we maximize the mutual information between local features and high-level features of their respective class. For training, we provide solely unlabeled images and start from random network initialization. For quantitative and qualitative evaluation, we use established benchmarks, and COCO-Persons, whereby we introduce the latter in this paper as a challenging novel benchmark. InfoSeg significantly outperforms the current state-of-the-art, e.g., we achieve a relative increase of 26% in the Pixel Accuracy metric on the COCO-Stuff dataset. © 2021, CC BY-SA.

关键词： Image representation

End-to-end Trainable Deep Neural Network for Robotic Grasp Detection and Semantic Segmentation from RGB

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Ainetter, Stefan Fraundorfer, Friedrich The Institute of Computer Graphics and Vision Graz University of Technology Graz8010 Austria

In this work, we introduce a novel, end-to-end trainable CNN-based architecture to deliver high quality results for grasp detection suitable for a parallel-plate gripper, and semantic segmentation. Utilizing this, we propose a novel refinement module that takes advantage of previously calculated grasp detection and semantic segmentation and further increases grasp detection accuracy. Our proposed network delivers state-of-the-art accuracy on two popular grasp dataset, namely Cornell and Jacquard. As additional contribution, we provide a novel dataset extension for the OCID dataset, making it possible to evaluate grasp detection in highly challenging scenes. Using this dataset, we show that semantic segmentation can additionally be used to assign grasp candidates to object classes, which can be used to pick specific objects in the scene. Source code and dataset extension are available at https://***/stefan-ainetter/grasp_det_seg_cnn. Copyright © 2021, The Authors. All rights reserved.

关键词： Semantic Segmentation

Video Test-Time Adaptation for Action Recognition

学校读者我要写书评

暂无评论

Video Test-Time Adaptation for Action Recognition

Conference on computer vision and Pattern Recognition (CVPR)

作者： Wei Lin Muhammad Jehanzeb Mirza Mateusz Kozinski Horst Possegger Hilde Kuehne Horst Bischof Institute for Computer Graphics and Vision Graz University of Technology Austria Christian Doppler Laboratory for Semantic 3D Computer Vision Christian Doppler Laboratory for Embedded Machine Learning Goethe University Frankfurt Germany MIT-IBM Watson AI Lab

Although action recognition systems can achieve top performance when evaluated on in-distribution test points, they are vulnerable to unanticipated distribution shifts in test data. However, test-time adaptation of video action recognition models against common distribution shifts has so far not been demonstrated. We propose to address this problem with an approach tailored to spatio-temporal models that is capable of adaptation on a single video sample at a step. It consists in a feature distribution alignment technique that aligns online estimates of test set statistics towards the training statistics. We further enforce prediction consistency over temporally augmented views of the same test video sample. Evaluations on three benchmark action recognition datasets show that our proposed technique is architecture-agnostic and able to significantly boost the performance on both, the state of the art convolutional architecture TANet and the Video Swin Transformer. Our proposed method demonstrates a substantial performance gain over existing test-time adaptation approaches in both evaluations of a single distribution shift and the challenging case of random distribution shifts. Code will be available at https://***/wlin-at/ViTTA.

关键词：

Benchmarking Classical and Learning-Based Multibeam Point Cloud Registration

学校读者我要写书评

暂无评论

Benchmarking Classical and Learning-Based Multibeam Point Cl...

IEEE International Conference on Robotics and Automation (ICRA)

作者： Li Ling Jun Zhang Nils Bore John Folkesson Anna Wåhlin Division of Robotics Perception and Learning (RPL) KTH Royal Institute of Technology Stockholm Sweden Institute of Computer Graphics and Vision (ICGV) TU Graz Austria Ocean Infinity Västra Frölunda Sweden Department of Marine Sciences University of Gothenburg Sweden

ISBN: (数字)9798350384574

ISBN: (纸本)9798350384581

Deep learning has shown promising results for multiple 3D point cloud registration datasets. However, in the underwater domain, most registration of multibeam echo-sounder (MBES) point cloud data are still performed using classical methods in the iterative closest point (ICP) family. In this work, we curate and release DotsonEast Dataset, a semi-synthetic MBES registration dataset constructed from an autonomous underwater vehicle in West Antarctica. Using this dataset, we systematically benchmark the performance of 2 classical and 4 learning-based methods. The experimental results show that the learning-based methods work well for coarse alignment, and are better at recovering rough transforms consistently at high overlap (20-50%). In comparison, GICP (a variant of ICP) performs well for fine alignment and is better across all metrics at extremely low overlap (10%). To the best of our knowledge, this is the first work to benchmark both learning-based and classical registration methods on an AUV-based MBES dataset. To facilitate future research, both the code and data are made available online. 1

关键词： Point cloud compression Learning systems Measurement Three-dimensional displays Simultaneous localization and mapping Iterative closest point algorithm Transforms