Depth Map Estimation Using Defocus and Motion Cues

Kumar, Himanshu; Yadav, Ajeet Singh; Gupta, Sumana; Venkatesh, K. S.

doi:10.1109/tcsvt.2018.2832086

Cited by 17 publications

(3 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[30]- [39] adopted different strategies to obtain reliable depth estimation from a single camera by learning to exploit monocular clues such as shadows, occlusions and relative scales between objects. In this field, a particularly appealing practice consists of training end-to-end models in a self or semisupervised manner [12], [36], replacing the need for groundtruth depth labels with image reprojection across different viewpoints according to two main strategies, respectively acquiring images with a single, moving camera [36], [38], [40], [41] or using a stereo camera [10], [12], [37], [42]- [44].…”

Section: A Depth Estimationmentioning

confidence: 99%

Monocular Depth Perception on Microcontrollers for Edge Applications

Peluso

Cipolletta

Calimera

et al. 2022

IEEE Trans. Circuits Syst. Video Technol.

View full text Add to dashboard Cite

Depth estimation is crucial in several computer vision applications, and a recent trend in this field aims at inferring such a cue from a single camera. Unfortunately, despite the compelling results achieved, state-of-the-art monocular depth estimation methods are computationally demanding, thus precluding their practical deployment in several application contexts characterized by low-power constraints. Therefore, in this paper, we propose a lightweight Convolutional Neural Network based on a shallow pyramidal architecture, referred to as µPyD-Net, enabling monocular depth estimation on microcontrollers. The network is trained in a peculiar self-supervised manner leveraging proxy labels obtained through a traditional stereo algorithm. Moreover, we propose optimization strategies aimed at performing computations with quantized 8-bit data and map the high-level description of the network to low-level layers optimized for the target microcontroller architecture. Exhaustive experimental results on standard datasets and an in-depth evaluation with a device belonging to the popular Arm Cortex-M family confirm that obtaining sufficiently accurate monocular depth estimation on microcontrollers is feasible. To the best of our knowledge, our proposal is the first one enabling such remarkable achievement, paving the way for the deployment of monocular depth cues onto the tiny end-nodes of distributed sensor networks.

show abstract

Section: A Depth Estimationmentioning

confidence: 99%

Monocular Depth Perception on Microcontrollers for Edge Applications

Peluso

Cipolletta

Calimera

et al. 2022

IEEE Trans. Circuits Syst. Video Technol.

View full text Add to dashboard Cite

show abstract

“…Also, estimated defocus value results in value proportional to original defocus value due to effect such as downsampling as shown in (11), matting [54] required for full defocus map estimation from sparse defocus map, etc. We use optics to relate defocus parameter

σ

to depth d as given by (13) based on framework discussed in [56]. Here,

k_{1}

and

k_{2}

are camera parameters that are always positive

σ = k_{1} - \frac{k_{2}}{d} \Rightarrow d = \frac{k_{2}}{k_{1} - σ}

Estimation of defocus is relative in nature, i.e.…”

Section: Applications Of Bplcmentioning

confidence: 99%

Blur parameter locus curve and its applications

Kumar¹,

Gupta²,

Subramanian³

2020

IET Image Processing

Self Cite

View full text Add to dashboard Cite

“…In contrast to learning-based schemes, traditional depth estimation methods mainly focus on discovering different cues to estimate depth from 2-D images [1]- [4], such as structure, shadows, lighting, occlusion, etc [5]. However, these cues are limited or only applicable to specific scenarios (e.g., shadows are greatly affected by ambient lighting).…”

Section: Introductionmentioning

confidence: 99%

Neural Contourlet Network for Monocular 360° Depth Estimation

Shen

Lin

Nie

et al. 2022

IEEE Trans. Circuits Syst. Video Technol.

View full text Add to dashboard Cite

For a monocular 360°image, depth estimation is a challenging because the distortion increases along the latitude. To perceive the distortion, existing methods devote to designing a deep and complex network architecture. In this paper, we provide a new perspective that constructs an interpretable and sparse representation for a 360°image. Considering the importance of the geometric structure in depth estimation, we utilize the contourlet transform to capture an explicit geometric cue in the spectral domain and integrate it with an implicit cue in the spatial domain. Specifically, we propose a neural contourlet network consisting of a convolutional neural network and a contourlet transform branch. In the encoder stage, we design a spatial-spectral fusion module to effectively fuse two types of cues. Contrary to the encoder, we employ the inverse contourlet transform with learned low-pass subbands and band-pass directional subbands to compose the depth in the decoder. Experiments on the three popular 360°panoramic image datasets demonstrate that the proposed approach outperforms the state-of-the-art schemes with faster convergence. Code is available at https://github.com/ zhijieshen-bjtu/Neural-Contourlet-Network-for-MODE.

show abstract

Depth Map Estimation Using Defocus and Motion Cues

Cited by 17 publications

References 31 publications

Monocular Depth Perception on Microcontrollers for Edge Applications

Monocular Depth Perception on Microcontrollers for Edge Applications

Blur parameter locus curve and its applications

Neural Contourlet Network for Monocular 360° Depth Estimation

Contact Info

Product

Resources

About