Matching in the Dark: A Dataset for Matching Image Pairs of Low-light Scenes

Song, Wenzheng; Suganuma, Masanori; Liu, Xing; Shimobayashi, Noriyuki; Maruta, Daisuke; Okatani, Takayuki

doi:10.1109/iccv48922.2021.00597

Cited by 13 publications

(5 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to suppress the adverse impact of cross-domain adaptation (CDA) in [31] equation (3), this paper proposes a modified CDA mechanism to adjust the conditional probability distribution [32]. The class conditional probability distribution can be expressed as:…”

Section: Improved Joint Allocation Adaptabilitymentioning

confidence: 99%

Wind turbines fault diagnosis method under variable working conditions based on AMVMD and deep discrimination transfer learning network

Shi,

Jia,

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

With the wide application of wind turbines, the bearing fault diagnosis of wind turbines has become a research hotspot. Under complex variable working conditions, the vibration signals of bearing components show non-stationary characteristics. Therefore, it is challenging to extract fault features using typical fault diagnosis methods. This paper proposes Adaptive Multivariate Variational Mode Decomposition combined with an improved Deep Discrimination Transfer Learning Network (AMVMD-IDDTLN) for bearing fault diagnosis of wind turbines under variable working conditions. First, the AMVMD method is used for the adaptive decomposition of the original signal, and use SE-ResNet18 convolutional neural network to obtain the transfer features of the source domain and target domain. Then, marginal distribution differences and conditional differences are assessed by DDM measures. The whole model is optimized by cross-entropy and IJDA loss function, and the identification and classification of cross-working fault characteristics of the wind turbine- bearings are realized. The model achieves 99.48% transfer learning for the ten classifications of CWRU data set, 97% transfer learning for the four classifications of UPB data set, and 90% transfer learning for wind turbine bearing data across working conditions and across equipment. It is concluded that: Compared with similar models, the AMVMD-IDDTLN model proposed in this paper has higher diagnostic accuracy and faster convergence rate, which has certain practicality.

show abstract

Section: Improved Joint Allocation Adaptabilitymentioning

confidence: 99%

Wind turbines fault diagnosis method under variable working conditions based on AMVMD and deep discrimination transfer learning network

Shi,

Jia,

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…On the route of increasing the number of image nodes, IMGpedia [13] is the largest graph which contains 14,765,300 image nodes. Nevertheless, IMGpedia is more like a vision-similaritybased image library [45,70] with image descriptions and meta information, it lacks the most valuable feature of the knowledge graph: "The Logical Connection".…”

Section: Multimodel Knowledge Graphmentioning

confidence: 99%

UKnow: A Unified Knowledge Protocol for Common-Sense Reasoning and Vision-Language Pre-training

Gong¹,

Xie²,

Feng³

et al. 2023

Preprint

View full text Add to dashboard Cite

This work presents a unified knowledge protocol, called UKnow, which facilitates knowledge-based studies from the perspective of data. Particularly focusing on visual and linguistic modalities, we categorize data knowledge into five unit types, namely, in-image, intext, cross-image, cross-text, and image-text. Following this protocol, we collect, from public international news, a large-scale multimodal knowledge graph dataset that consists of 1,388,568 nodes (with 571,791 visionrelated ones) and 3,673,817 triplets. The dataset is also annotated with rich event tags, including 96 coarse labels and 9,185 fine labels, expanding its potential usage. To further verify that UKnow can serve as a standard protocol, we set up an efficient pipeline to help reorganize existing datasets under UKnow format. Finally, we benchmark the performance of some widely-used baselines on the tasks of common-sense reasoning and visionlanguage pre-training. Results on both our new dataset and the reformatted public datasets demonstrate the effectiveness of UKnow in knowledge organization and method evaluation. Code, dataset, conversion tool, and baseline models will be made public. (https://github. com/Gongggg/UKnow)

show abstract

“…We evaluate the proposed method on the challenging Matching In the Dark (MID) benchmark (Song et al 2021), a largescale low-light stereo RAW image dataset, which is also the only currently available dataset for low-light image matching evaluation. The MID dataset contains 54 indoor scenes and 54 outdoor scenes, and each scene contains 48 stereo image pairs in RAW format taken with 6 different exposure times and 8 different ISOs.…”

Section: Experiments Dataset and Metricsmentioning

confidence: 99%

“…Although a considerable measure of noise still exists, RAW images have larger bit widths and retain richer original information than RGB images, which are processed by the image signal processing (ISP) module within cameras. Recently, based on RAW images, a benchmark called MID (Song et al 2021) with a denoise-then-detect pipeline is proposed, which first denoises RAW images using BM3D (Dabov et al 2007) or SID (Chen et al 2018) and then applies feature detection, description and matching algorithms. Nevertheless, existing denoising algorithms are usually very time-consuming and sometimes introduce artifacts that hinder accurate keypoint detection, which limit the application of this denoise-then-detect pipeline.…”

Section: Introductionmentioning

confidence: 99%

DarkFeat: Noise-Robust Feature Detector and Descriptor for Extremely Low-Light RAW Images

Wang

et al. 2023

AAAI

View full text Add to dashboard Cite

Low-light visual perception, such as SLAM or SfM at night, has received increasing attention, in which keypoint detection and local feature description play an important role. Both handcraft designs and machine learning methods have been widely studied for local feature detection and description, however, the performance of existing methods degrades in the extreme low-light scenarios in a certain degree, due to the low signal-to-noise ratio in images. To address this challenge, images in RAW format that retain more raw sensing information have been considered in recent works with a denoise-then-detect scheme. However, existing denoising methods are still insufficient for RAW images and heavily time-consuming, which limits the practical applications of such scheme. In this paper, we propose DarkFeat, a deep learning model which directly detects and describes local features from extreme low-light RAW images in an end-to-end manner. A novel noise robustness map and selective suppression constraints are proposed to effectively mitigate the influence of noise and extract more reliable keypoints. Furthermore, a customized pipeline of synthesizing dataset containing low-light RAW image matching pairs is proposed to extend end-to-end training. Experimental results show that DarkFeat achieves state-of-the-art performance on both indoor and outdoor parts of the challenging MID benchmark, outperforms the denoise-then-detect methods and significantly reduces computational costs up to 70%. Code is available at https://github.com/THU-LYJ-Lab/DarkFeat.

show abstract

Matching in the Dark: A Dataset for Matching Image Pairs of Low-light Scenes

Cited by 13 publications

References 44 publications

Wind turbines fault diagnosis method under variable working conditions based on AMVMD and deep discrimination transfer learning network

Wind turbines fault diagnosis method under variable working conditions based on AMVMD and deep discrimination transfer learning network

UKnow: A Unified Knowledge Protocol for Common-Sense Reasoning and Vision-Language Pre-training

DarkFeat: Noise-Robust Feature Detector and Descriptor for Extremely Low-Light RAW Images

Contact Info

Product

Resources

About