3,448 Hits in 6.5 sec

Simultaneous Semantic Segmentation and Outlier Detection in Presence of Domain Shift [article]

Petra Bevandić, Ivan Krešo, Marin Oršić, Siniša Šegvić
2019 arXiv   pre-print
We consider semantic segmentation as the primary task and perform extensive validation on WildDash val (inliers), LSUN val (outliers), and pasted objects from Pascal VOC 2007 (outliers).  ...  We evaluate our best two models on the WildDash test dataset and set a new state of the art on the WildDash benchmark.  ...  Simultaneous Segmentation and Outlier Detection Our method combines two distinct tasks: outlier detection and semantic segmentation, as shown in Fig. 1 .  ... 
arXiv:1908.01098v1 fatcat:rtqijtnmx5hitfzwdn4jrjgdwe

A Survey of Visual Sensory Anomaly Detection [article]

Xi Jiang, Guoyang Xie, Jinbao Wang, Yong Liu, Chengjie Wang, Feng Zheng, Yaochu Jin
2022 arXiv   pre-print
Compared with semantic anomaly detection which detects anomaly at the label level (semantic shift), visual sensory AD detects the abnormal part of the sample (covariate shift).  ...  In this survey, we are the first one to provide a comprehensive review of visual sensory AD and category into three levels according to the form of anomalies.  ...  Therefore, the author innovatively detects novel categories in semantic segmentation and retrieves their semantic similarity to enhance the domain shift capability of the model.  ... 
arXiv:2202.07006v1 fatcat:2bqzmmrnjzggti5tcewa3mh3sa

FCNs in the Wild: Pixel-level Adversarial and Constraint-based Adaptation [article]

Judy Hoffman, Dequan Wang, Fisher Yu, Trevor Darrell
2016 arXiv   pre-print
In this paper, we introduce the first domain adaptive semantic segmentation method, proposing an unsupervised adversarial approach to pixel prediction problems.  ...  Global domain alignment is performed using a novel semantic segmentation network with fully convolutional domain adversarial learning.  ...  Related Work Semantic Segmentation Semantic segmentation is a key computer vision task and has been studied in a plethora of publications.  ... 
arXiv:1612.02649v1 fatcat:3ku35eodhvcwxpboh4hrggnq3y

Anomaly Detection in Autonomous Driving: A Survey [article]

Daniel Bogdoll, Maximilian Nitsche, J. Marius Zöllner
2022 arXiv   pre-print
This survey provides an extensive overview of anomaly detection techniques based on camera, lidar, radar, multimodal and abstract object level data.  ...  We outline the state-of-the-art and point out current research gaps.  ...  Acknowledgment This work results from the project KI Data Tooling (19A20001J), funded by the German Federal Ministry for Economic Affairs and Climate Action (BMWK).  ... 
arXiv:2204.07974v1 fatcat:3rdola4tjfesllr6dqhqgf3tve

Continual Active Learning Using Pseudo-Domains for Limited Labelling Resources and Changing Acquisition Characteristics [article]

Matthias Perkonigg, Johannes Hofmanninger, Christian Herold, Helmut Prosch, Georg Langs
2022 arXiv   pre-print
To demonstrate generalizability, we evaluate the effectiveness of our method on three tasks: cardiac segmentation, lung nodule detection and brain age estimation.  ...  The approach automatically recognizes shifts in image acquisition characteristics - new domains -, selects optimal examples for labelling and adapts training accordingly.  ...  Ethical Standards The work follows appropriate ethical standards in conducting research and writing the manuscript, following all applicable laws and regulations regarding treatment of animals or human  ... 
arXiv:2111.13069v2 fatcat:jlejzfxpkbd3raxllazwppmcfe

Anomaly detection in information streams without prior domain knowledge

M. S. Beigi, S.-F. Chang, S. Ebadollahi, D. C. Verma
2011 IBM Journal of Research and Development  
The proposed approach simultaneously monitors and analyzes the data stream at multiple temporal scales and learns the evolution of normal behavior over time in each time scale.  ...  The proposed approach is not sensitive to the choice of the distance metric and hence is applicable in various domains and applications.  ...  The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of the U.S.  ... 
doi:10.1147/jrd.2011.2163280 fatcat:khgej2lkqrcynetkjgqitu46si

Detecting and Learning the Unknown in Semantic Segmentation [article]

Robin Chan, Svenja Uhlemeyer, Matthias Rottmann, Hanno Gottschalk
2022 arXiv   pre-print
Deep neural networks (DNNs) are commonly used for this task and they are usually trained on a closed set of object classes appearing in a closed operational domain.  ...  In this work, we first give an overview about anomalies from an information-theoretic perspective. Next, we review research in detecting semantically unknown objects in semantic segmentation.  ...  ", grant no. 19A19005R, and "KI Delta Learning -Scalable AI for Automated Driving", grant no. 19A19013Q, respectively.  ... 
arXiv:2202.08700v1 fatcat:6e6vvviq2zed5idpuv3iuqre7i

Inspect, Understand, Overcome: A Survey of Practical Methods for AI Safety [article]

Sebastian Houben, Stephanie Abrecht, Maram Akila, Andreas Bär, Felix Brockherde, Patrick Feifel, Tim Fingscheidt, Sujan Sai Gannamaneni, Seyed Eghbal Ghobadi, Ahmed Hammam, Anselm Haselhoff, Felix Hauser (+29 others)
2021 arXiv   pre-print
In recent years, a zoo of state-of-the-art techniques aiming to address these safety concerns has emerged. This work provides a structured and broad overview of them.  ...  The use of deep neural networks (DNNs) in safety-critical applications like mobile health and autonomous driving is challenging due to numerous model-inherent shortcomings.  ...  Furthermore, this research has been funded by the Federal Ministry of Education and Research of Germany as part of the competence center for machine learning ML2R (01IS18038B).  ... 
arXiv:2104.14235v1 fatcat:f6sj3v2brza7thyzw7b7fkpo2m

Multimodal Detection of Unknown Objects on Roads for Autonomous Driving [article]

Daniel Bogdoll and Enrico Eisen and Maximilian Nitsche and Christin Scheib and J. Marius Zöllner
2022 arXiv   pre-print
Instead of focusing on a single sensor modality, we make use of lidar and camera data by combining state-of-the art detection models in a sequential manner.  ...  We evaluate our approach on the Waymo Open Perception Dataset and point out current research gaps in anomaly detection.  ...  Road Segmentation In order to focus primarily on objects on the road, we applied a semantic segmentation model in the image domain as a first step.  ... 
arXiv:2205.01414v3 fatcat:3gfhymmchffanbkfklg47224nm

Text Driven Temporal Segmentation of Cricket Videos [chapter]

K. Pramod Sankar, Saurabh Pandey, C. V. Jawahar
2006 Lecture Notes in Computer Science  
This allows for a semantic access and retrieval of video segments, which is difficult to obtain from existing visual feature based approaches.  ...  Specifically, we segment videos based on textual descriptions or commentaries of the action in the video.  ...  . • Semantic access to content, has met with much success in the text retrieval domain.  ... 
doi:10.1007/11949619_39 fatcat:3vsyza4n4rgtvcyr4mpdvxetkm

Track, then Decide: Category-Agnostic Vision-based Multi-Object Tracking [article]

Aljoša Ošep, Wolfgang Mehner, Paul Voigtlaender, Bastian Leibe
2017 arXiv   pre-print
Our approach can utilize semantic information whenever it is available for classifying objects at the track level, while retaining the capability to track generic unknown objects in the absence of such  ...  The most common paradigm for vision-based multi-object tracking is tracking-by-detection, due to the availability of reliable detectors for several important object categories such as cars and pedestrians  ...  Inference At this stage, we obtain several conflicting segmentation hypotheses in the time domain, including over-and undersegmentations, as well as outliers from the static background, such as segments  ... 
arXiv:1712.07920v1 fatcat:2jbolkg7k5fq3kvprzza3jfw54

Taxonomy of Machine Learning Safety: A Survey and Primer [article]

Sina Mohseni and Haotao Wang and Zhiding Yu and Chaowei Xiao and Zhangyang Wang and Jay Yadawa
2022 arXiv   pre-print
Research explores different approaches to improve ML dependability by proposing new models and training techniques to reduce generalization error, achieve domain adaptation, and detect outlier examples  ...  In this paper, we present a structured and comprehensive review of ML techniques to improve the dependability of ML algorithms in uncontrolled open-world settings.  ...  [132] constructed a composite semantic segmentation dataset from multiple sources to improve the zero-shot generalization ability to unseen domains of semantic segmentation models.  ... 
arXiv:2106.04823v2 fatcat:mpw6v2j5mnbvho4w6tfhu7umeu

Indoor Building Reconstruction from Occluded Point Clouds Using Graph-Cut and Ray-Tracing

Mattia Previtali, Lucía Díaz-Vilariño, Marco Scaioni
2018 Applied Sciences  
Specific challenges are related to the complex building layouts and the high presence of elements such as pieces of furniture causing clutter and occlusions.  ...  Doors and windows are detected from occlusions by implementing a ray-tracing algorithm. The methodology is tested in a real case study.  ...  Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/app8091529 fatcat:zarvrujqr5hmjf2yaksyujjtfa

Single-Image Piece-Wise Planar 3D Reconstruction via Associative Embedding

Zehao Yu, Jia Zheng, Dongze Lian, Zihan Zhou, Shenghua Gao
2019 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
Single-image piece-wise planar 3D reconstruction aims to simultaneously segment plane instances and recover 3D plane parameters from an image.  ...  With the proposed method, we are able to detect an arbitrary number of planes. Extensive experiments on public datasets validate the effectiveness and efficiency of our method.  ...  conditions) and outliers (e.g., due to the presence of non-planar objects).  ... 
doi:10.1109/cvpr.2019.00112 dblp:conf/cvpr/YuZLZG19 fatcat:ei34v24qkrgnbeep6drhbupgje

Semantic Video Analysis Based on Estimation and Representation of Higher-Order Motion Statistics

Georgios Th. Papadopoulos, Alexia Briassouli, Vasileios Mezaris, Ioannis Kompatsiaris, Michael G. Strintzis
2008 2008 Third International Workshop on Semantic Media Adaptation and Personalization  
Then, Hidden Markov Models (HMMs) are employed for performing the association of each shot with one of the semantic classes that are of interest in any given domain.  ...  Experimental results as well as comparative evaluation from the application of the proposed approach in the domain of news broadcast video are presented.  ...  This is because they appear as outliers, and in [10] it is proven that the kurtosis is a robust, locally optimum test statistic, for the detection of outliers, even in the presence of non-Gaussian noise  ... 
doi:10.1109/smap.2008.22 dblp:conf/smap/PapadopoulosBMKS08 fatcat:dpfxtwseenakta5j2vhxxfth7a
« Previous Showing results 1 — 15 out of 3,448 results