Related papers: Multiple data sources and domain generalization learning method for road surface defect classification

Multiple data sources and domain generalization learning method for road surface defect classification

URL: http://arxiv.org/abs/2407.10197v1
Date: Sun, 14 Jul 2024 13:37:47 GMT
Title: Multiple data sources and domain generalization learning method for road surface defect classification
Authors: Linh Trinh, Ali Anwar, Siegfried Mercelis,
Abstract summary: We propose a method for classifying road surface defects using camera images. We present a domain generalization training algorithm for developing a generalized model. The results show that our method can efficiently classify road surface defects on previously unseen data.
Score: 2.9109581496560044
License: http://creativecommons.org/licenses/by-nc-sa/4.0/
Abstract: Roads are an essential mode of transportation, and maintaining them is critical to economic growth and citizen well-being. With the continued advancement of AI, road surface inspection based on camera images has recently been extensively researched and can be performed automatically. However, because almost all of the deep learning methods for detecting road surface defects were optimized for a specific dataset, they are difficult to apply to a new, previously unseen dataset. Furthermore, there is a lack of research on training an efficient model using multiple data sources. In this paper, we propose a method for classifying road surface defects using camera images. In our method, we propose a scheme for dealing with the invariance of multiple data sources while training a model on multiple data sources. Furthermore, we present a domain generalization training algorithm for developing a generalized model that can work with new, completely unseen data sources without requiring model updates. We validate our method using an experiment with six data sources corresponding to six countries from the RDD2022 dataset. The results show that our method can efficiently classify road surface defects on previously unseen data.

Related papers

UdeerLID+: Integrating LiDAR, Image, and Relative Depth with Semi-Supervised [12.440461420762265]
Road segmentation is a critical task for autonomous driving systems. Our work introduces an innovative approach that integrates LiDAR point cloud data, visual image, and relative depth maps. One of the primary challenges is the scarcity of large-scale, accurately labeled datasets.
arXiv Detail & Related papers (2024-09-10T03:57:30Z)
Self-supervised Feature Adaptation for 3D Industrial Anomaly Detection [59.41026558455904]
We focus on multi-modal anomaly detection. Specifically, we investigate early multi-modal approaches that attempted to utilize models pre-trained on large-scale visual datasets. We propose a Local-to-global Self-supervised Feature Adaptation (LSFA) method to finetune the adaptors and learn task-oriented representation toward anomaly detection.
arXiv Detail & Related papers (2024-01-06T07:30:41Z)
Exploring Data Redundancy in Real-world Image Classification through Data Selection [20.389636181891515]
Deep learning models often require large amounts of data for training, leading to increased costs. We present two data valuation metrics based on Synaptic Intelligence and gradient norms, respectively, to study redundancy in real-world image data. Online and offline data selection algorithms are then proposed via clustering and grouping based on the examined data values.
arXiv Detail & Related papers (2023-06-25T03:31:05Z)
Exploring the Effectiveness of Dataset Synthesis: An application of Apple Detection in Orchards [68.95806641664713]
We explore the usability of Stable Diffusion 2.1-base for generating synthetic datasets of apple trees for object detection. We train a YOLOv5m object detection model to predict apples in a real-world apple detection dataset. Results demonstrate that the model trained on generated data is slightly underperforming compared to a baseline model trained on real-world images.
arXiv Detail & Related papers (2023-06-20T09:46:01Z)
Minimizing the Accumulated Trajectory Error to Improve Dataset Distillation [151.70234052015948]
We propose a novel approach that encourages the optimization algorithm to seek a flat trajectory. We show that the weights trained on synthetic data are robust against the accumulated errors perturbations with the regularization towards the flat trajectory. Our method, called Flat Trajectory Distillation (FTD), is shown to boost the performance of gradient-matching methods by up to 4.7%.
arXiv Detail & Related papers (2022-11-20T15:49:11Z)
One-Pass Learning via Bridging Orthogonal Gradient Descent and Recursive Least-Squares [8.443742714362521]
We develop an algorithm for one-pass learning which seeks to perfectly fit every new datapoint while changing the parameters in a direction that causes the least change to the predictions on previous datapoints. Our algorithm uses the memory efficiently by exploiting the structure of the streaming data via an incremental principal component analysis (IPCA) Our experiments show the effectiveness of the proposed method compared to the baselines.
arXiv Detail & Related papers (2022-07-28T02:01:31Z)
What Stops Learning-based 3D Registration from Working in the Real World? [53.68326201131434]
This work identifies the sources of 3D point cloud registration failures, analyze the reasons behind them, and propose solutions. Ultimately, this translates to a best-practice 3D registration network (BPNet), constituting the first learning-based method able to handle previously-unseen objects in real-world data. Our model generalizes to real data without any fine-tuning, reaching an accuracy of up to 67% on point clouds of unseen objects obtained with a commercial sensor.
arXiv Detail & Related papers (2021-11-19T19:24:27Z)
Hybrid Model and Data Driven Algorithm for Online Learning of Any-to-Any Path Loss Maps [19.963385352536616]
Learning any-to-any path loss maps might be a key enabler for applications that rely on device-to-any (D2D) communication. Model-based methods have the advantage that they can generate reliable estimations with low computational complexity. Pure data-driven methods can achieve good performance without assuming any physical model. We propose a novel hybrid model and data-driven approach that obtained datasets from an online fashion.
arXiv Detail & Related papers (2021-07-14T13:08:25Z)
Unsupervised Model Drift Estimation with Batch Normalization Statistics for Dataset Shift Detection and Model Selection [0.0]
We propose a novel method of model drift estimation by exploiting statistics of batch normalization layer on unlabeled test data. We show the effectiveness of our method not only on dataset shift detection but also on model selection when there are multiple candidate models among model zoo or training trajectories in an unsupervised way.
arXiv Detail & Related papers (2021-07-01T03:04:47Z)
Unsupervised Multi-source Domain Adaptation Without Access to Source Data [58.551861130011886]
Unsupervised Domain Adaptation (UDA) aims to learn a predictor model for an unlabeled domain by transferring knowledge from a separate labeled source domain. We propose a novel and efficient algorithm which automatically combines the source models with suitable weights in such a way that it performs at least as good as the best source model.
arXiv Detail & Related papers (2021-04-05T10:45:12Z)
Deep Traffic Sign Detection and Recognition Without Target Domain Real Images [52.079665469286496]
We propose a novel database generation method that requires no real image from the target-domain, and (ii) templates of the traffic signs. The method does not aim at overcoming the training with real data, but to be a compatible alternative when the real data is not available. On large data sets, training with a fully synthetic data set almost matches the performance of training with a real one.
arXiv Detail & Related papers (2020-07-30T21:06:47Z)

This list is automatically generated from the titles and abstracts of the papers in this site.