Monocular Segment-Wise Depth: Monocular Depth Estimation Based on a Semantic Segmentation Prior

Atapour-Abarghouei, Amir; Breckon, Toby P.

doi:10.1109/icip.2019.8803551

Cited by 11 publications

(12 citation statements)

References 19 publications

(47 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The CNN-based methods are more accurate than the SVM method in [ 9 , 10 ], but the accuracy is still not sufficient enough to realize secure flight without collisions. In [ 11 , 23 ], the authors proposed methods to preprocess segmenting images before depth estimation using a monocular camera. This method improves the accuracy of depth estimation.…”

Section: Related Workmentioning

confidence: 99%

“…On the other hand, the computational workload largely increases, and the method is not suitable for real-time processing in terms of the performance since the methods [ 11 , 23 ] need segmentation as pre-process. In addition, the methods [ 11 , 19 , 20 , 21 , 22 , 23 ] use public datasets [ 24 , 25 ] to estimate depth. As the latest work, AdaBins based on a transformer has been proposed [ 26 ].…”

Section: Related Workmentioning

confidence: 99%

“…Many kinds of research for autonomous flight of drones have assumed that monocular cameras are often used to detect and recognize objects around the drones [ 7 , 8 ]. Single monocular camera-based depth estimation is also actively researched [ 9 , 10 , 11 , 12 ]. However, monocular cameras are not useful at night in terms of their visibility.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Pix2Pix-Based Monocular Depth Estimation for Drones with Optical Flow on AirSim

Shimada

Nishikawa

Kong

et al. 2022

Sensors

View full text Add to dashboard Cite

In this work, we propose a method for estimating depth for an image of a monocular camera in order to avoid a collision for the autonomous flight of a drone. The highest flight speed of a drone is generally approximate 22.2 m/s, and long-distant depth information is crucial for autonomous flights since if the long-distance information is not available, the drone flying at high speeds is prone to collisions. However, long-range, measurable depth cameras are too heavy to be equipped on a drone. This work applies Pix2Pix, which is a kind of Conditional Generative Adversarial Nets (CGAN). Pix2Pix generates depth images from a monocular camera. Additionally, this work applies optical flow to enhance the accuracy of depth estimation. In this work, we propose a highly accurate depth estimation method that effectively embeds an optical flow map into a monocular image. The models are trained with taking advantage of AirSim, which is one of the flight simulators. AirSim can take both monocular and depth images over a hundred meter in the virtual environment, and our model generates a depth image that provides the long-distance information than images captured by a common depth camera. We evaluate accuracy and error of our proposed method using test images in AirSim. In addition, the proposed method is utilized for flight simulation to evaluate the effectiveness to collision avoidance. As a result, our proposed method is higher accuracy and lower error than a state of work. Moreover, our proposed method is lower collision than a state of work.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Pix2Pix-Based Monocular Depth Estimation for Drones with Optical Flow on AirSim

Shimada

Nishikawa

Kong

et al. 2022

Sensors

View full text Add to dashboard Cite

show abstract

“…The ever-increasing number of data breaches and security attacks observed on a regular basis [17] and the innovative use of novel attacks by cybercriminals emphasise the importance of getting ahead of the curve using cutting-edge techniques such as AI. Modern artificial intelligence and machine learning approaches [18]- [27] have revolutionised numerous areas of research with significant industrial, societal and economic impact. Making use of such AI-based methods to predict and prevent breaches and attacks would give the cybersecurity industry the advantage they urgently need.…”

Section: What Is the Issue With Data And Sharing It?mentioning

confidence: 99%

Resolving the cybersecurity Data Sharing Paradox to scale up cybersecurity via a co-production approach towards data sharing

Atapour-Abarghouei¹,

McGough²,

Wall³

2020

Preprint

Self Cite

View full text Add to dashboard Cite

As cybercriminals scale up their operations to increase their profits or inflict greater harm, we argue that there is an equal need to respond to their threats by scaling up cybersecurity. To achieve this goal, we have to develop a co-productive approach towards data collection and sharing by overcoming the cybersecurity data sharing paradox. This is where we all agree on the definition of the problem and end goal (improving cybersecurity and getting rid of cybercrime), but we disagree about how to achieve it and fail to work together efficiently. At the core of this paradox is the observation that public interests differ from private interests. As a result, industry and law enforcement take different approaches to the cybersecurity problem as they seek to resolve incidents in their own interests, which manifests in different data sharing practices between both and also other interested parties, such as cybersecurity researchers. The big question we ask is can these interests be reconciled to develop an interdisciplinary approach towards co-operation and sharing data. In essence, all three will have to co-own the problem in order to co-produce a solution. We argue that a few operational models with good practices exist that provide guides to a possible solution, especially multiple third-party ownership organisations which consolidate, anonymise and analyse data. To take this forward, we suggest the practical solution of organising coproductive data collection on a sectoral basis, but acknowledge that common standards for data collection will also have to be developed and agreed upon. We propose an initial set of best practices for building collaborations and sharing data and argue that these best practices need to be developed and standardised in order to mitigate the paradox.

show abstract

“…In recent years, monocular depth estimation (i.e. estimating scene depth from a single RGB image) has received widespread attention within both academia and industry as a more effective, economical and innovative alternative to more conventional depth estimation strategies [5,6,14,18,20,31,66].…”

Section: Introductionmentioning

confidence: 99%

To Complete or to Estimate, That is the Question: A Multi-Task Approach to Depth Completion and Monocular Depth Estimation

Atapour-Abarghouei

Breckon

2019

2019 International Conference on 3D Vision (3DV)

Self Cite

View full text Add to dashboard Cite

Robust three-dimensional scene understanding is now an ever-growing area of research highly relevant in many realworld applications such as autonomous driving and robotic navigation. In this paper, we propose a multi-task learningbased model capable of performing two tasks:-sparse depth completion (i.e. generating complete dense scene depth given a sparse depth image as the input) and monocular depth estimation (i.e. predicting scene depth from a single RGB image) via two sub-networks jointly trained end to end using data randomly sampled from a publicly available corpus of synthetic and real-world images. The first subnetwork generates a sparse depth image by learning lower level features from the scene and the second predicts a full dense depth image of the entire scene, leading to a better geometric and contextual understanding of the scene and, as a result, superior performance of the approach. The entire model can be used to infer complete scene depth from a single RGB image or the second network can be used alone to perform depth completion given a sparse depth input. Using adversarial training, a robust objective function, a deep architecture relying on skip connections and a blend of synthetic and real-world training data, our approach is capable of producing superior high quality scene depth. Extensive experimental evaluation demonstrates the efficacy of our approach compared to contemporary state-of-the-art techniques across both problem domains.

show abstract

Monocular Segment-Wise Depth: Monocular Depth Estimation Based on a Semantic Segmentation Prior

Cited by 11 publications

References 19 publications

Pix2Pix-Based Monocular Depth Estimation for Drones with Optical Flow on AirSim

Pix2Pix-Based Monocular Depth Estimation for Drones with Optical Flow on AirSim

Resolving the cybersecurity Data Sharing Paradox to scale up cybersecurity via a co-production approach towards data sharing

To Complete or to Estimate, That is the Question: A Multi-Task Approach to Depth Completion and Monocular Depth Estimation

Contact Info

Product

Resources

About