Single image calibration using knowledge distillation approaches
- URL: http://arxiv.org/abs/2212.02379v1
- Date: Mon, 5 Dec 2022 15:59:35 GMT
- Title: Single image calibration using knowledge distillation approaches
- Authors: Khadidja Ould Amer, Oussama Hadjerci, Mohamed Abbas Hedjazi, Antoine
Letienne
- Abstract summary: We build upon a CNN architecture to automatically estimate camera parameters.
We adapt four common incremental learning strategies to preserve knowledge when updating the network for new data distributions.
Experiment results were significant and indicated which method was better for the camera calibration estimation.
- Score: 1.7205106391379026
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Although recent deep learning-based calibration methods can predict extrinsic
and intrinsic camera parameters from a single image, their generalization
remains limited by the number and distribution of training data samples. The
huge computational and space requirement prevents convolutional neural networks
(CNNs) from being implemented in resource-constrained environments. This
challenge motivated us to learn a CNN gradually, by training new data while
maintaining performance on previously learned data. Our approach builds upon a
CNN architecture to automatically estimate camera parameters (focal length,
pitch, and roll) using different incremental learning strategies to preserve
knowledge when updating the network for new data distributions. Precisely, we
adapt four common incremental learning, namely: LwF , iCaRL, LU CIR, and BiC by
modifying their loss functions to our regression problem. We evaluate on two
datasets containing 299008 indoor and outdoor images. Experiment results were
significant and indicated which method was better for the camera calibration
estimation.
Related papers
- Data-efficient Event Camera Pre-training via Disentangled Masked
Modeling [20.987277885575963]
We present a new data-supervised voxel-based self-supervised learning method for event cameras.
Our method overcomes the limitations of previous methods, which either sacrifice temporal information or directly employ paired image data.
It exhibits excellent generalization performance and demonstrates significant improvements across various tasks with fewer parameters and lower computational costs.
arXiv Detail & Related papers (2024-03-01T10:02:25Z) - Leveraging Neural Radiance Fields for Uncertainty-Aware Visual
Localization [56.95046107046027]
We propose to leverage Neural Radiance Fields (NeRF) to generate training samples for scene coordinate regression.
Despite NeRF's efficiency in rendering, many of the rendered data are polluted by artifacts or only contain minimal information gain.
arXiv Detail & Related papers (2023-10-10T20:11:13Z) - Deep Multi-Threshold Spiking-UNet for Image Processing [51.88730892920031]
This paper introduces the novel concept of Spiking-UNet for image processing, which combines the power of Spiking Neural Networks (SNNs) with the U-Net architecture.
To achieve an efficient Spiking-UNet, we face two primary challenges: ensuring high-fidelity information propagation through the network via spikes and formulating an effective training strategy.
Experimental results show that, on image segmentation and denoising, our Spiking-UNet achieves comparable performance to its non-spiking counterpart.
arXiv Detail & Related papers (2023-07-20T16:00:19Z) - Learning Markerless Robot-Depth Camera Calibration and End-Effector Pose
Estimation [0.0]
We present a learning-based markerless extrinsic calibration system that uses a depth camera and does not rely on simulation data.
We learn models for end-effector (EE) segmentation, single-frame rotation prediction and keypoint detection, from automatically generated real-world data.
Our robustness with training data from multiple camera poses and test data from previously unseen poses give sub-centimeter evaluations and sub-deciradian average calibration and pose estimation errors.
arXiv Detail & Related papers (2022-12-15T00:53:42Z) - Multi-task Learning for Camera Calibration [3.274290296343038]
We present a unique method for predicting intrinsic (principal point offset and focal length) and extrinsic (baseline, pitch, and translation) properties from a pair of images.
By reconstructing the 3D points using a camera model neural network and then using the loss in reconstruction to obtain the camera specifications, this innovative camera projection loss (CPL) method allows us that the desired parameters should be estimated.
arXiv Detail & Related papers (2022-11-22T17:39:31Z) - Agricultural Plantation Classification using Transfer Learning Approach
based on CNN [0.0]
The efficiency of hyper-spectral image recognition has increased significantly with deep learning.
CNN and Multi-Layer Perceptron(MLP) has demonstrated to be an excellent process of classifying images.
We propose using the method of transfer learning to decrease the training time and reduce the dependence on large labeled data-set.
arXiv Detail & Related papers (2022-06-19T14:43:31Z) - Image Restoration by Deep Projected GSURE [115.57142046076164]
Ill-posed inverse problems appear in many image processing applications, such as deblurring and super-resolution.
We propose a new image restoration framework that is based on minimizing a loss function that includes a "projected-version" of the Generalized SteinUnbiased Risk Estimator (GSURE) and parameterization of the latent image by a CNN.
arXiv Detail & Related papers (2021-02-04T08:52:46Z) - RIFLE: Backpropagation in Depth for Deep Transfer Learning through
Re-Initializing the Fully-connected LayEr [60.07531696857743]
Fine-tuning the deep convolution neural network(CNN) using a pre-trained model helps transfer knowledge learned from larger datasets to the target task.
We propose RIFLE - a strategy that deepens backpropagation in transfer learning settings.
RIFLE brings meaningful updates to the weights of deep CNN layers and improves low-level feature learning.
arXiv Detail & Related papers (2020-07-07T11:27:43Z) - Curriculum By Smoothing [52.08553521577014]
Convolutional Neural Networks (CNNs) have shown impressive performance in computer vision tasks such as image classification, detection, and segmentation.
We propose an elegant curriculum based scheme that smoothes the feature embedding of a CNN using anti-aliasing or low-pass filters.
As the amount of information in the feature maps increases during training, the network is able to progressively learn better representations of the data.
arXiv Detail & Related papers (2020-03-03T07:27:44Z) - Calibrating Deep Neural Networks using Focal Loss [77.92765139898906]
Miscalibration is a mismatch between a model's confidence and its correctness.
We show that focal loss allows us to learn models that are already very well calibrated.
We show that our approach achieves state-of-the-art calibration without compromising on accuracy in almost all cases.
arXiv Detail & Related papers (2020-02-21T17:35:50Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.