Guided Stereo Matching

Poggi, Matteo; Pallotti, Davide; Tosi, Fabio; Mattoccia, Stefano

doi:10.1109/cvpr.2019.00107

Cited by 76 publications

(77 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Batsos et al [2] soften this effect by combining traditional functions and confidence measures [15,31] within a random forest framework, proving better generalization compared to CNN-based method [44]. Finally, guiding end-to-end CNNs with external depth measurements (e.g.Lidar) allows for reducing the domain-shift effect, as reported in [30]. Image reconstruction for unsupervised learning.…”

Section: Related Workmentioning

confidence: 99%

Real-Time Self-Adaptive Deep Stereo

Tonioni

Tosi

Poggi

et al. 2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

Self Cite

260

228

View full text Add to dashboard Cite

Deep convolutional neural networks trained end-to-end are the state-of-the-art methods to regress dense disparity maps from stereo pairs. These models, however, suffer from a notable decrease in accuracy when exposed to scenarios significantly different from the training set (e.g., real vs synthetic images, etc.). We argue that it is extremely unlikely to gather enough samples to achieve effective training/tuning in any target domain, thus making this setup impractical for many applications. Instead, we propose to perform unsupervised and continuous online adaptation of a deep stereo network, which allows for preserving its accuracy in any environment. However, this strategy is extremely computationally demanding and thus prevents real-time inference. We address this issue introducing a new lightweight, yet effective, deep stereo architecture, Modularly ADaptive Network (MADNet), and developing a Modular ADaptation (MAD) algorithm, which independently trains sub-portions of the network.By deploying MADNet together with MAD we introduce the first real-time self-adaptive deep stereo system enabling competitive performance on heterogeneous datasets.Our code is publicly available at https://github.com/CVLAB-Unibo/ Real-time-self-adaptive-deep-stereo.

show abstract

Section: Related Workmentioning

confidence: 99%

Real-Time Self-Adaptive Deep Stereo

Tonioni

Tosi

Poggi

et al. 2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

Self Cite

260

228

View full text Add to dashboard Cite

show abstract

“…In contrast to our work, (Poggi et al 2019) uses disparities by sparse LiDAR points to filter cost volume during both training and testing. Both AcfNet and the method of (Poggi et al 2019) are trained on Scene Flow from scratch, and directly evaluated on training sets of KITTI 2012 and 2015 since (Poggi et al 2019) requires sparse LiDAR points as inputs. Table 3 reports the comparison results, where AcfNet outperforms (Poggi et al 2019) on all performance metrics by large margins even without using LiDAR points as inputs.…”

Section: Cost Volume Filtering Comparisonsmentioning

confidence: 96%

“…To further validate the superiority of the proposed cost volume filtering, experiments are designed to compare with the concurrent work (Poggi et al 2019). In contrast to our work, (Poggi et al 2019) uses disparities by sparse LiDAR points to filter cost volume during both training and testing. Both AcfNet and the method of (Poggi et al 2019) are trained on Scene Flow from scratch, and directly evaluated on training sets of KITTI 2012 and 2015 since (Poggi et al 2019) requires sparse LiDAR points as inputs.…”

Section: Cost Volume Filtering Comparisonsmentioning

confidence: 99%

Adaptive Unimodal Cost Volume Filtering for Deep Stereo Matching

Zhang

Chen

Bai

et al. 2020

AAAI

138

View full text Add to dashboard Cite

State-of-the-art deep learning based stereo matching approaches treat disparity estimation as a regression problem, where loss function is directly defined on true disparities and their estimated ones. However, disparity is just a byproduct of a matching process modeled by cost volume, while indirectly learning cost volume driven by disparity regression is prone to overfitting since the cost volume is under constrained. In this paper, we propose to directly add constraints to the cost volume by filtering cost volume with unimodal distribution peaked at true disparities. In addition, variances of the unimodal distributions for each pixel are estimated to explicitly model matching uncertainty under different contexts. The proposed architecture achieves state-of-the-art performance on Scene Flow and two KITTI stereo benchmarks. In particular, our method ranked the 1st place of KITTI 2012 evaluation and the 4th place of KITTI 2015 evaluation (recorded on 2019.8.20). The codes of AcfNet are available at: https://github.com/youmi-zym/AcfNet.

show abstract

“…synthetic to real) has been addressed in either offline [49] or online [50] fashion, or greatly reduced by guiding them with external depth measurements (e.g. Lidar) [42]. Monocular depth estimation.…”

Section: Related Workmentioning

confidence: 99%

Learning Monocular Depth Estimation Infusing Traditional Stereo Knowledge

Tosi

Aleotti

Poggi

et al. 2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

Self Cite

202

167

View full text Add to dashboard Cite

Depth estimation from a single image represents a fascinating, yet challenging problem with countless applications. Recent works proved that this task could be learned without direct supervision from ground truth labels leveraging image synthesis on sequences or stereo pairs. Focusing on this second case, in this paper we leverage stereo matching in order to improve monocular depth estimation. To this aim we propose monoResMatch, a novel deep architecture designed to infer depth from a single input image by synthesizing features from a different point of view, horizontally aligned with the input image, performing stereo matching between the two cues. In contrast to previous works sharing this rationale, our network is the first trained end-to-end from scratch. Moreover, we show how obtaining proxy ground truth annotation through traditional stereo algorithms, such as Semi-Global Matching, enables more accurate monocular depth estimation still countering the need for expensive depth labels by keeping a self-supervised approach. Exhaustive experimental results prove how the synergy between i) the proposed monoResMatch architecture and ii) proxy-supervision attains state-of-theart for self-supervised monocular depth estimation. The code is publicly available at https

show abstract

Guided Stereo Matching

Cited by 76 publications

References 41 publications

Real-Time Self-Adaptive Deep Stereo

Real-Time Self-Adaptive Deep Stereo

Adaptive Unimodal Cost Volume Filtering for Deep Stereo Matching

Learning Monocular Depth Estimation Infusing Traditional Stereo Knowledge

Contact Info

Product

Resources

About