Deep Direct Visual Odometry

Zhao, Chaoqiang; Tang, Yang; Sun, Qiyu

doi:10.1109/tits.2021.3071886

Cited by 21 publications

(12 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As mentioned above concerning the resilient application of the proposed approach, the equivalent workflow could be adapted for more data such as trajectory evaluation [35]. Furthermore, the approach can be useful for data privacy [2,36] to identify the safe and unsafe regions.…”

Section: Resultsmentioning

confidence: 99%

Coupling algebraic topology theory, formal methods and safety requirements toward a new coverage metric for artificial intelligence models

Adjed

Mziou-Sallami

Pelliccia

et al. 2022

Neural Comput & Applic

View full text Add to dashboard Cite

HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L'archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d'enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

show abstract

Section: Resultsmentioning

confidence: 99%

Coupling algebraic topology theory, formal methods and safety requirements toward a new coverage metric for artificial intelligence models

Adjed

Mziou-Sallami

Pelliccia

et al. 2022

Neural Comput & Applic

View full text Add to dashboard Cite

show abstract

“…Therefore, the Euclidean distance between nodes i and j , for a given route, can be calculated as:

where

is the velocity of the neighbor vehicle j compared to previous node i . Alternatively, inter-distance between nodes can be calculated as in [ 34 ]. Hence, the best position

between one node and its adjacent l node in a given route out of k TMR routes is:

The

is calculated with the maximum distance between the source node and the nodes in a given TMR route to consider the closest node to the destination vehicle/RSU.…”

Section: Methodsmentioning

confidence: 99%

Data Dissemination in VANETs Using Particle Swarm Optimization

Desai

El-Ocla

Purohit

2023

Sensors

View full text Add to dashboard Cite

A vehicular Ad-Hoc Network (VANET) is a type of Mobile Ad-Hoc Networks (MANETs) that uses wireless routers inside each vehicle to act as a node. The need for effective solutions to urban traffic congestion issues has increased recently due to the growing number of automobile connections in the car communications system. To ensure a high level of service and avoid unsafe situations brought on by congestion or a broadcast storm, data dissemination in a VANET network requires an effective approach. Effective multi-objective optimization methods are required to tackle this because of the implied competing nature of multi-metric approaches. A meta-heuristic technique with a high level of solution interactions can handle efficient optimization. To accomplish this, a meta-heuristic search algorithm particle optimization was chosen. In this paper, we have created a network consisting of vehicles as nodes. The aim is to send emergency messages immediately to the stationary nodes. The normal messages will be sent to the FIFO queue. To send these messages to a destination node, multiple routes were found using Time delay-based Multipath Routing (TMR) method, and to find the optimal and secure path Particle Swarm Optimization (PSO) is used. Our method is compared with different optimization methods such as Ant Colony Optimization (ACO), Firefly Optimization (FFO), and Enhanced Flying Ant Colony Optimization (EFACO). Significant improvements in terms of throughput and packet loss ratio, reduced end-to-end delay, rounding overhead ratio, and the energy consumption are revealed by the experimental results.

show abstract

“…Later, the same research group that developed LSD-SLAM proposed DSO, which brought significant improvements in monocular odometry by optimizing camera poses, intrinsic and map parameters, boosting results through the use of photometric camera calibration. Subsequently, several versions of this work were presented, mainly addressing the use of different camera types [ 3 , 23 ], a modification with loop closure [ 24 ] and using predictions provided by deep learning techniques [ 25 ]. Yang et al [ 26 ] proposed a semi-supervised network called StackNet for deep monocular depth estimation; these estimates were incorporated to overcome scale drift, an intrinsic limitation in monocular VO.…”

Section: Related Workmentioning

confidence: 99%

DPO: Direct Planar Odometry with Stereo Camera

Lins

Rosa

Grassi

et al. 2023

Sensors

View full text Add to dashboard Cite

Nowadays, state-of-the-art direct visual odometry (VO) methods essentially rely on points to estimate the pose of the camera and reconstruct the environment. Direct Sparse Odometry (DSO) became the standard technique and many approaches have been developed from it. However, only recently, two monocular plane-based DSOs have been presented. The first one uses a learning-based plane estimator to generate coarse planes as input for optimization. When these coarse estimates are too far from the minimum, the optimization may fail. Thus, the entire system result is dependent on the quality of the plane predictions and restricted to the training data domain. The second one only detects planes in vertical and horizontal orientation as being more adequate to structured environments. To the best of our knowledge, we propose the first Stereo Plane-based VO inspired by the DSO framework. Differing from the above-mentioned methods, our approach purely uses planes as features in the sliding window optimization and uses a dual quaternion as pose parameterization. The conducted experiments showed that our method presents a similar performance to Stereo DSO, a point-based approach.

show abstract

Deep Direct Visual Odometry

Cited by 21 publications

References 30 publications

Coupling algebraic topology theory, formal methods and safety requirements toward a new coverage metric for artificial intelligence models

Coupling algebraic topology theory, formal methods and safety requirements toward a new coverage metric for artificial intelligence models

Data Dissemination in VANETs Using Particle Swarm Optimization

DPO: Direct Planar Odometry with Stereo Camera

Contact Info

Product

Resources

About