Unsupervised Learning of Optical Flow With Non-Occlusion From Geometry

Wang, Guangming; Ren, Shuaiqi; Wang, Hesheng

doi:10.1109/tits.2022.3182858

Cited by 13 publications

(2 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The aforementioned three aspects support our research intuition to develop a transformer backbone for LiDAR odometry. Motivated by the motion information estimation method from soft correspondences in optical flow (Wang, Ren, and Wang 2022) and scene flow tasks (Wang et al 2022a(Wang et al ,b, 2021a, we directly estimate pose from conditioned features after the cross attention. Overall, our contributions are as follows:…”

Section: Introductionmentioning

confidence: 99%

TransLO: A Window-Based Masked Point Transformer Framework for Large-Scale LiDAR Odometry

Liu

Wang

Chaokang

et al. 2023

AAAI

View full text Add to dashboard Cite

Recently, transformer architecture has gained great success in the computer vision community, such as image classification, object detection, etc. Nonetheless, its application for 3D vision remains to be explored, given that point cloud is inherently sparse, irregular, and unordered. Furthermore, existing point transformer frameworks usually feed raw point cloud of N×3 dimension into transformers, which limits the point processing scale because of their quadratic computational costs to the input size N. In this paper, we rethink the structure of point transformer. Instead of directly applying transformer to points, our network (TransLO) can process tens of thousands of points simultaneously by projecting points onto a 2D surface and then feeding them into a local transformer with linear complexity. Specifically, it is mainly composed of two components: Window-based Masked transformer with Self Attention (WMSA) to capture long-range dependencies; Masked Cross-Frame Attention (MCFA) to associate two frames and predict pose estimation. To deal with the sparsity issue of point cloud, we propose a binary mask to remove invalid and dynamic points. To our knowledge, this is the first transformer-based LiDAR odometry network. The experiment results on the KITTI odometry dataset show that our average rotation and translation RMSE achieves 0.500°/100m and 0.993% respectively. The performance of our network surpasses all recent learning-based methods and even outperforms LOAM on most evaluation sequences.Codes will be released on https://github.com/IRMVLab/TransLO.

show abstract

Section: Introductionmentioning

confidence: 99%

TransLO: A Window-Based Masked Point Transformer Framework for Large-Scale LiDAR Odometry

Liu

Wang

Chaokang

et al. 2023

AAAI

View full text Add to dashboard Cite

show abstract

“…Sedangkan, Unsupervised learning adalah teknik yang tidak menggunakan data yang telah diberi label atau kategori oleh manusia. Dalam teknik ini, mesin dapat mencari pola dan hubungan dalam data tanpa adanya panduan atau kategori tertentu [9].…”

unclassified

Analisis Manfaat Machine Learning Pada Next-Generation Firewall Sophos Xg 330 Dalam Mengatasi Serangan SQL Injection

Putra¹,

Pratama²,

Sutabri³

2023

misi

View full text Add to dashboard Cite

Dengan perkembangan dunia digital yang terus berkembang, penggunaan teknologi informasi menghasilkan sejumlah besar data yang berharga, dengan adanya data yang berharga ini membuat perusahaan yang menyimpannya memiliki risiko terhadap serangan siber. Salah satu ancaman siber yang umum terjadi adalah SQL Injection yang dapat memanipulasi hak akses dan menyebabkan kebocoran data, kehilangan data, dan kehilangan hak akses ke dalam database. Sehingga membuat perusahaan dan organisasi untuk dapat melindungi data mereka dari segala jenis ancaman yang mungkin akan terjadi. Dalam upaya untuk dapat mengatasi serangan SQL Injection penggunaan teknologi Next-Generation Firewall (NGFW) seperti Sophos XG 330 telah muncul sebagai solusi yang lebih unggul dibandingkan dengan firewall tradisional. NGFW mampu melakukan inspeksi paket yang lebih rinci dan menerapkan teknik pembelajaran mesin, memungkinkan mereka untuk mendeteksi ancaman keamanan yang lebih kompleks atau serangan yang sebelumnya tidak diketahui. Dalam penelitian ini kami melakukan analisis terkait dengan manfaat penerapan dari machine learning pada Next-Generation Firewall, dalam hal untuk dapat mengidentifikasi dan memblokir serangan yang terjadi. Hasil dari penelitian ini memberikan pengetahuan untuk perusahaan atau organisasi dalam memilih solusi yang tepat untuk dapat melindungi data mereka.

show abstract

Unsupervised learning of optical flow in a multi-frame dynamic environment using temporal dynamic modeling

Sun,

Luo,

Nishida

2023

Complex Intell. Syst.

View full text Add to dashboard Cite

For visual estimation of optical flow, which is crucial for various vision analyses, unsupervised learning by view synthesis has emerged as a promising alternative to supervised methods because the ground-truth flow is not readily available in many cases. However, unsupervised learning is likely to be unstable when pixel tracking is lost via occlusion and motion blur, or pixel correspondence is impaired by variations in image content and spatial structure over time. Recognizing that dynamic occlusions and object variations usually exhibit a smooth temporal transition in natural settings, we shifted our focus to model unsupervised learning optical flow from multi-frame sequences of such dynamic scenes. Specifically, we simulated various dynamic scenarios and occlusion phenomena based on Markov property, allowing the model to extract motion laws and thus gain performance in dynamic and occluded areas, which diverges from existing methods without considering temporal dynamics. In addition, we introduced a temporal dynamic model based on a well-designed spatial-temporal dual recurrent block, resulting in a lightweight model structure with fast inference speed. Assuming the temporal smoothness of optical flow, we used the prior motions of adjacent frames to supervise the occluded regions more reliably. Experiments on several optical flow benchmarks demonstrated the effectiveness of our method, as the performance is comparable to several state-of-the-art methods with advantages in memory and computational overhead.

show abstract

Unsupervised Learning of Optical Flow With Non-Occlusion From Geometry

Cited by 13 publications

References 37 publications

TransLO: A Window-Based Masked Point Transformer Framework for Large-Scale LiDAR Odometry

TransLO: A Window-Based Masked Point Transformer Framework for Large-Scale LiDAR Odometry

Analisis Manfaat Machine Learning Pada Next-Generation Firewall Sophos Xg 330 Dalam Mengatasi Serangan SQL Injection

Unsupervised learning of optical flow in a multi-frame dynamic environment using temporal dynamic modeling

Contact Info

Product

Resources

About