Pub Date : 2022-09-29DOI: 10.1007/978-3-031-19821-2_28
Alexander C. Li, Alexei A. Efros, Deepak Pathak
{"title":"Understanding Collapse in Non-contrastive Siamese Representation Learning","authors":"Alexander C. Li, Alexei A. Efros, Deepak Pathak","doi":"10.1007/978-3-031-19821-2_28","DOIUrl":"https://doi.org/10.1007/978-3-031-19821-2_28","url":null,"abstract":"","PeriodicalId":72676,"journal":{"name":"Computer vision - ECCV ... : ... European Conference on Computer Vision : proceedings. European Conference on Computer Vision","volume":"28 1","pages":"490-505"},"PeriodicalIF":0.0,"publicationDate":"2022-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73642709","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-09-27DOI: 10.48550/arXiv.2209.13362
Yijin Li, Xinyang Liu, Wenqian Dong, Han Zhou, H. Bao, Guofeng Zhang, Yinda Zhang, Zhaopeng Cui
Light-weight time-of-flight (ToF) depth sensors are small, cheap, low-energy and have been massively deployed on mobile devices for the purposes like autofocus, obstacle detection, etc. However, due to their specific measurements (depth distribution in a region instead of the depth value at a certain pixel) and extremely low resolution, they are insufficient for applications requiring high-fidelity depth such as 3D reconstruction. In this paper, we propose DELTAR, a novel method to empower light-weight ToF sensors with the capability of measuring high resolution and accurate depth by cooperating with a color image. As the core of DELTAR, a feature extractor customized for depth distribution and an attention-based neural architecture is proposed to fuse the information from the color and ToF domain efficiently. To evaluate our system in real-world scenarios, we design a data collection device and propose a new approach to calibrate the RGB camera and ToF sensor. Experiments show that our method produces more accurate depth than existing frameworks designed for depth completion and depth super-resolution and achieves on par performance with a commodity-level RGB-D sensor. Code and data are available at https://zju3dv.github.io/deltar/.
{"title":"DELTAR: Depth Estimation from a Light-weight ToF Sensor and RGB Image","authors":"Yijin Li, Xinyang Liu, Wenqian Dong, Han Zhou, H. Bao, Guofeng Zhang, Yinda Zhang, Zhaopeng Cui","doi":"10.48550/arXiv.2209.13362","DOIUrl":"https://doi.org/10.48550/arXiv.2209.13362","url":null,"abstract":"Light-weight time-of-flight (ToF) depth sensors are small, cheap, low-energy and have been massively deployed on mobile devices for the purposes like autofocus, obstacle detection, etc. However, due to their specific measurements (depth distribution in a region instead of the depth value at a certain pixel) and extremely low resolution, they are insufficient for applications requiring high-fidelity depth such as 3D reconstruction. In this paper, we propose DELTAR, a novel method to empower light-weight ToF sensors with the capability of measuring high resolution and accurate depth by cooperating with a color image. As the core of DELTAR, a feature extractor customized for depth distribution and an attention-based neural architecture is proposed to fuse the information from the color and ToF domain efficiently. To evaluate our system in real-world scenarios, we design a data collection device and propose a new approach to calibrate the RGB camera and ToF sensor. Experiments show that our method produces more accurate depth than existing frameworks designed for depth completion and depth super-resolution and achieves on par performance with a commodity-level RGB-D sensor. Code and data are available at https://zju3dv.github.io/deltar/.","PeriodicalId":72676,"journal":{"name":"Computer vision - ECCV ... : ... European Conference on Computer Vision : proceedings. European Conference on Computer Vision","volume":"15 1","pages":"619-636"},"PeriodicalIF":0.0,"publicationDate":"2022-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83874951","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Under the domain shift, cross-domain few-shot object detection aims to adapt object detectors in the target domain with a few annotated target data. There exists two significant challenges: (1) Highly insufficient target domain data; (2) Potential over-adaptation and misleading caused by inappropriately amplified target samples without any restriction. To address these challenges, we propose an adaptive method consisting of two parts. First, we propose an adaptive optimization strategy to select augmented data similar to target samples rather than blindly increasing the amount. Specifically, we filter the augmented candidates which significantly deviate from the target feature distribution in the very beginning. Second, to further relieve the data limitation, we propose the multi-level domain-aware data augmentation to increase the diversity and rationality of augmented data, which exploits the cross-image foreground-background mixture. Experiments show that the proposed method achieves state-of-the-art performance on multiple benchmarks.
{"title":"AcroFOD: An Adaptive Method for Cross-domain Few-shot Object Detection","authors":"Yipeng Gao, Lingxiao Yang, Yunmu Huang, Song Xie, Shiyong Li, Weihao Zheng","doi":"10.48550/arXiv.2209.10904","DOIUrl":"https://doi.org/10.48550/arXiv.2209.10904","url":null,"abstract":"Under the domain shift, cross-domain few-shot object detection aims to adapt object detectors in the target domain with a few annotated target data. There exists two significant challenges: (1) Highly insufficient target domain data; (2) Potential over-adaptation and misleading caused by inappropriately amplified target samples without any restriction. To address these challenges, we propose an adaptive method consisting of two parts. First, we propose an adaptive optimization strategy to select augmented data similar to target samples rather than blindly increasing the amount. Specifically, we filter the augmented candidates which significantly deviate from the target feature distribution in the very beginning. Second, to further relieve the data limitation, we propose the multi-level domain-aware data augmentation to increase the diversity and rationality of augmented data, which exploits the cross-image foreground-background mixture. Experiments show that the proposed method achieves state-of-the-art performance on multiple benchmarks.","PeriodicalId":72676,"journal":{"name":"Computer vision - ECCV ... : ... European Conference on Computer Vision : proceedings. European Conference on Computer Vision","volume":"285 1","pages":"673-690"},"PeriodicalIF":0.0,"publicationDate":"2022-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77736103","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-09-22DOI: 10.48550/arXiv.2209.11277
Fabian Duffhauss, Ngo Anh Vien, Hanna Ziesche, G. Neumann
. Sensor fusion can significantly improve the performance of many computer vision tasks. However, traditional fusion approaches are either not data-driven and cannot exploit prior knowledge nor find regu-larities in a given dataset or they are restricted to a single application. We overcome this shortcoming by presenting a novel deep hierarchical variational autoencoder called FusionVAE that can serve as a basis for many fusion tasks. Our approach is able to generate diverse image samples that are conditioned on multiple noisy, occluded, or only partially visible input images. We derive and optimize a variational lower bound for the conditional log-likelihood of FusionVAE. In order to assess the fusion capabilities of our model thoroughly, we created three novel datasets for image fusion based on popular computer vision datasets. In our experiments, we show that FusionVAE learns a representation of aggregated information that is relevant to fusion tasks. The results demonstrate that our approach outperforms traditional methods significantly. Furthermore, we present the advantages and disadvantages of different design choices.
{"title":"FusionVAE: A Deep Hierarchical Variational Autoencoder for RGB Image Fusion","authors":"Fabian Duffhauss, Ngo Anh Vien, Hanna Ziesche, G. Neumann","doi":"10.48550/arXiv.2209.11277","DOIUrl":"https://doi.org/10.48550/arXiv.2209.11277","url":null,"abstract":". Sensor fusion can significantly improve the performance of many computer vision tasks. However, traditional fusion approaches are either not data-driven and cannot exploit prior knowledge nor find regu-larities in a given dataset or they are restricted to a single application. We overcome this shortcoming by presenting a novel deep hierarchical variational autoencoder called FusionVAE that can serve as a basis for many fusion tasks. Our approach is able to generate diverse image samples that are conditioned on multiple noisy, occluded, or only partially visible input images. We derive and optimize a variational lower bound for the conditional log-likelihood of FusionVAE. In order to assess the fusion capabilities of our model thoroughly, we created three novel datasets for image fusion based on popular computer vision datasets. In our experiments, we show that FusionVAE learns a representation of aggregated information that is relevant to fusion tasks. The results demonstrate that our approach outperforms traditional methods significantly. Furthermore, we present the advantages and disadvantages of different design choices.","PeriodicalId":72676,"journal":{"name":"Computer vision - ECCV ... : ... European Conference on Computer Vision : proceedings. European Conference on Computer Vision","volume":"36 1","pages":"674-691"},"PeriodicalIF":0.0,"publicationDate":"2022-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86771546","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-09-22DOI: 10.48550/arXiv.2209.10811
S. Moon, GyeongMoon Park
Recently, manipulation of real-world images has been highly elaborated along with the development of Generative Adversarial Networks (GANs) and corresponding encoders, which embed real-world images into the latent space. However, designing encoders of GAN still remains a challenging task due to the trade-off between distortion and perception. In this paper, we point out that the existing encoders try to lower the distortion not only on the interest region, e.g., human facial region but also on the uninterest region, e.g., background patterns and obstacles. However, most uninterest regions in real-world images are located at out-of-distribution (OOD), which are infeasible to be ideally reconstructed by generative models. Moreover, we empirically find that the uninterest region overlapped with the interest region can mangle the original feature of the interest region, e.g., a microphone overlapped with a facial region is inverted into the white beard. As a result, lowering the distortion of the whole image while maintaining the perceptual quality is very challenging. To overcome this trade-off, we propose a simple yet effective encoder training scheme, coined IntereStyle, which facilitates encoding by focusing on the interest region. IntereStyle steers the encoder to disentangle the encodings of the interest and uninterest regions. To this end, we filter the information of the uninterest region iteratively to regulate the negative impact of the uninterest region. We demonstrate that IntereStyle achieves both lower distortion and higher perceptual quality compared to the existing state-of-the-art encoders. Especially, our model robustly conserves features of the original images, which shows the robust image editing and style mixing results. We will release our code with the pre-trained model after the review.
最近,随着生成对抗网络(GANs)和相应编码器的发展,对真实世界图像的处理得到了高度的阐述,这些编码器将真实世界的图像嵌入到潜在空间中。然而,由于失真和感知之间的权衡,GAN编码器的设计仍然是一个具有挑战性的任务。在本文中,我们指出现有的编码器不仅试图降低兴趣区域(如人脸区域)的失真,而且还试图降低非兴趣区域(如背景图案和障碍物)的失真。然而,现实图像中大多数的无兴趣区域位于out- distribution (OOD),无法通过生成模型进行理想的重构。此外,我们的经验发现,与兴趣区域重叠的非兴趣区域会扭曲兴趣区域的原始特征,例如,与面部区域重叠的麦克风会被反转成白胡子。因此,在保持感知质量的同时降低整个图像的失真是非常具有挑战性的。为了克服这种权衡,我们提出了一种简单而有效的编码器训练方案,称为IntereStyle,它通过关注感兴趣区域来促进编码。兴趣模式引导编码器解开感兴趣和不感兴趣区域的编码。为此,我们对无兴趣区域的信息进行迭代过滤,以调节无兴趣区域的负面影响。我们证明,与现有的最先进的编码器相比,IntereStyle实现了更低的失真和更高的感知质量。特别是该模型对原始图像的特征进行了鲁棒性保存,显示了鲁棒性的图像编辑和样式混合效果。我们将在审查后发布带有预训练模型的代码。
{"title":"IntereStyle: Encoding an Interest Region for Robust StyleGAN Inversion","authors":"S. Moon, GyeongMoon Park","doi":"10.48550/arXiv.2209.10811","DOIUrl":"https://doi.org/10.48550/arXiv.2209.10811","url":null,"abstract":"Recently, manipulation of real-world images has been highly elaborated along with the development of Generative Adversarial Networks (GANs) and corresponding encoders, which embed real-world images into the latent space. However, designing encoders of GAN still remains a challenging task due to the trade-off between distortion and perception. In this paper, we point out that the existing encoders try to lower the distortion not only on the interest region, e.g., human facial region but also on the uninterest region, e.g., background patterns and obstacles. However, most uninterest regions in real-world images are located at out-of-distribution (OOD), which are infeasible to be ideally reconstructed by generative models. Moreover, we empirically find that the uninterest region overlapped with the interest region can mangle the original feature of the interest region, e.g., a microphone overlapped with a facial region is inverted into the white beard. As a result, lowering the distortion of the whole image while maintaining the perceptual quality is very challenging. To overcome this trade-off, we propose a simple yet effective encoder training scheme, coined IntereStyle, which facilitates encoding by focusing on the interest region. IntereStyle steers the encoder to disentangle the encodings of the interest and uninterest regions. To this end, we filter the information of the uninterest region iteratively to regulate the negative impact of the uninterest region. We demonstrate that IntereStyle achieves both lower distortion and higher perceptual quality compared to the existing state-of-the-art encoders. Especially, our model robustly conserves features of the original images, which shows the robust image editing and style mixing results. We will release our code with the pre-trained model after the review.","PeriodicalId":72676,"journal":{"name":"Computer vision - ECCV ... : ... European Conference on Computer Vision : proceedings. European Conference on Computer Vision","volume":"22 1","pages":"460-476"},"PeriodicalIF":0.0,"publicationDate":"2022-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83881888","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-09-21DOI: 10.48550/arXiv.2209.10691
Liangchen Song, Xuan Gong, Benjamin Planche, Meng Zheng, D. Doermann, Junsong Yuan, Terrence Chen, Ziyan Wu
Knowing the 3D motions in a dynamic scene is essential to many vision applications. Recent progress is mainly focused on estimating the activity of some specific elements like humans. In this paper, we leverage a neural motion field for estimating the motion of all points in a multiview setting. Modeling the motion from a dynamic scene with multiview data is challenging due to the ambiguities in points of similar color and points with time-varying color. We propose to regularize the estimated motion to be predictable. If the motion from previous frames is known, then the motion in the near future should be predictable. Therefore, we introduce a predictability regularization by first conditioning the estimated motion on latent embeddings, then by adopting a predictor network to enforce predictability on the embeddings. The proposed framework PREF (Predictability REgularized Fields) achieves on par or better results than state-of-the-art neural motion field-based dynamic scene representation methods, while requiring no prior knowledge of the scene.
{"title":"PREF: Predictability Regularized Neural Motion Fields","authors":"Liangchen Song, Xuan Gong, Benjamin Planche, Meng Zheng, D. Doermann, Junsong Yuan, Terrence Chen, Ziyan Wu","doi":"10.48550/arXiv.2209.10691","DOIUrl":"https://doi.org/10.48550/arXiv.2209.10691","url":null,"abstract":"Knowing the 3D motions in a dynamic scene is essential to many vision applications. Recent progress is mainly focused on estimating the activity of some specific elements like humans. In this paper, we leverage a neural motion field for estimating the motion of all points in a multiview setting. Modeling the motion from a dynamic scene with multiview data is challenging due to the ambiguities in points of similar color and points with time-varying color. We propose to regularize the estimated motion to be predictable. If the motion from previous frames is known, then the motion in the near future should be predictable. Therefore, we introduce a predictability regularization by first conditioning the estimated motion on latent embeddings, then by adopting a predictor network to enforce predictability on the embeddings. The proposed framework PREF (Predictability REgularized Fields) achieves on par or better results than state-of-the-art neural motion field-based dynamic scene representation methods, while requiring no prior knowledge of the scene.","PeriodicalId":72676,"journal":{"name":"Computer vision - ECCV ... : ... European Conference on Computer Vision : proceedings. European Conference on Computer Vision","volume":"99 1","pages":"664-681"},"PeriodicalIF":0.0,"publicationDate":"2022-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74675657","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-09-21DOI: 10.48550/arXiv.2209.10677
Eric C. Yeats, Frank Liu, David A. P. Womble, Hai Li
We present a self-supervised method to disentangle factors of variation in high-dimensional data that does not rely on prior knowledge of the underlying variation profile (e.g., no assumptions on the number or distribution of the individual latent variables to be extracted). In this method which we call NashAE, high-dimensional feature disentanglement is accomplished in the low-dimensional latent space of a standard autoencoder (AE) by promoting the discrepancy between each encoding element and information of the element recovered from all other encoding elements. Disentanglement is promoted efficiently by framing this as a minmax game between the AE and an ensemble of regression networks which each provide an estimate of an element conditioned on an observation of all other elements. We quantitatively compare our approach with leading disentanglement methods using existing disentanglement metrics. Furthermore, we show that NashAE has increased reliability and increased capacity to capture salient data characteristics in the learned latent representation.
{"title":"NashAE: Disentangling Representations through Adversarial Covariance Minimization","authors":"Eric C. Yeats, Frank Liu, David A. P. Womble, Hai Li","doi":"10.48550/arXiv.2209.10677","DOIUrl":"https://doi.org/10.48550/arXiv.2209.10677","url":null,"abstract":"We present a self-supervised method to disentangle factors of variation in high-dimensional data that does not rely on prior knowledge of the underlying variation profile (e.g., no assumptions on the number or distribution of the individual latent variables to be extracted). In this method which we call NashAE, high-dimensional feature disentanglement is accomplished in the low-dimensional latent space of a standard autoencoder (AE) by promoting the discrepancy between each encoding element and information of the element recovered from all other encoding elements. Disentanglement is promoted efficiently by framing this as a minmax game between the AE and an ensemble of regression networks which each provide an estimate of an element conditioned on an observation of all other elements. We quantitatively compare our approach with leading disentanglement methods using existing disentanglement metrics. Furthermore, we show that NashAE has increased reliability and increased capacity to capture salient data characteristics in the learned latent representation.","PeriodicalId":72676,"journal":{"name":"Computer vision - ECCV ... : ... European Conference on Computer Vision : proceedings. European Conference on Computer Vision","volume":"13 1","pages":"36-51"},"PeriodicalIF":0.0,"publicationDate":"2022-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85665731","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-09-21DOI: 10.48550/arXiv.2209.10305
J. Fu, Hong Wang, Qi Xie, Qian Zhao, Deyu Meng, Zongben Xu
Although current deep learning-based methods have gained promising performance in the blind single image super-resolution (SISR) task, most of them mainly focus on heuristically constructing diverse network architectures and put less emphasis on the explicit embedding of the physical generation mechanism between blur kernels and high-resolution (HR) images. To alleviate this issue, we propose a model-driven deep neural network, called KXNet, for blind SISR. Specifically, to solve the classical SISR model, we propose a simple-yet-effective iterative algorithm. Then by unfolding the involved iterative steps into the corresponding network module, we naturally construct the KXNet. The main specificity of the proposed KXNet is that the entire learning process is fully and explicitly integrated with the inherent physical mechanism underlying this SISR task. Thus, the learned blur kernel has clear physical patterns and the mutually iterative process between blur kernel and HR image can soundly guide the KXNet to be evolved in the right direction. Extensive experiments on synthetic and real data finely demonstrate the superior accuracy and generality of our method beyond the current representative state-of-the-art blind SISR methods. Code is available at: https://github.com/jiahong-fu/KXNet.
{"title":"KXNet: A Model-Driven Deep Neural Network for Blind Super-Resolution","authors":"J. Fu, Hong Wang, Qi Xie, Qian Zhao, Deyu Meng, Zongben Xu","doi":"10.48550/arXiv.2209.10305","DOIUrl":"https://doi.org/10.48550/arXiv.2209.10305","url":null,"abstract":"Although current deep learning-based methods have gained promising performance in the blind single image super-resolution (SISR) task, most of them mainly focus on heuristically constructing diverse network architectures and put less emphasis on the explicit embedding of the physical generation mechanism between blur kernels and high-resolution (HR) images. To alleviate this issue, we propose a model-driven deep neural network, called KXNet, for blind SISR. Specifically, to solve the classical SISR model, we propose a simple-yet-effective iterative algorithm. Then by unfolding the involved iterative steps into the corresponding network module, we naturally construct the KXNet. The main specificity of the proposed KXNet is that the entire learning process is fully and explicitly integrated with the inherent physical mechanism underlying this SISR task. Thus, the learned blur kernel has clear physical patterns and the mutually iterative process between blur kernel and HR image can soundly guide the KXNet to be evolved in the right direction. Extensive experiments on synthetic and real data finely demonstrate the superior accuracy and generality of our method beyond the current representative state-of-the-art blind SISR methods. Code is available at: https://github.com/jiahong-fu/KXNet.","PeriodicalId":72676,"journal":{"name":"Computer vision - ECCV ... : ... European Conference on Computer Vision : proceedings. European Conference on Computer Vision","volume":"61 1","pages":"235-253"},"PeriodicalIF":0.0,"publicationDate":"2022-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86085056","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-09-19DOI: 10.48550/arXiv.2209.08924
Haoxian Zhang, Yonggen Ling
Robust and accurate planar tracking over a whole video sequence is vitally important for many vision applications. The key to planar object tracking is to find object correspondences, modeled by homography, between the reference image and the tracked image. Existing methods tend to obtain wrong correspondences with changing appearance variations, camera-object relative motions and occlusions. To alleviate this problem, we present a unified convolutional neural network (CNN) model that jointly considers homography, visibility, and confidence. First, we introduce correlation blocks that explicitly account for the local appearance changes and camera-object relative motions as the base of our model. Second, we jointly learn the homography and visibility that links camera-object relative motions with occlusions. Third, we propose a confidence module that actively monitors the estimation quality from the pixel correlation distributions obtained in correlation blocks. All these modules are plugged into a Lucas-Kanade (LK) tracking pipeline to obtain both accurate and robust planar object tracking. Our approach outperforms the state-of-the-art methods on public POT and TMT datasets. Its superior performance is also verified on a real-world application, synthesizing high-quality in-video advertisements.
{"title":"HVC-Net: Unifying Homography, Visibility, and Confidence Learning for Planar Object Tracking","authors":"Haoxian Zhang, Yonggen Ling","doi":"10.48550/arXiv.2209.08924","DOIUrl":"https://doi.org/10.48550/arXiv.2209.08924","url":null,"abstract":"Robust and accurate planar tracking over a whole video sequence is vitally important for many vision applications. The key to planar object tracking is to find object correspondences, modeled by homography, between the reference image and the tracked image. Existing methods tend to obtain wrong correspondences with changing appearance variations, camera-object relative motions and occlusions. To alleviate this problem, we present a unified convolutional neural network (CNN) model that jointly considers homography, visibility, and confidence. First, we introduce correlation blocks that explicitly account for the local appearance changes and camera-object relative motions as the base of our model. Second, we jointly learn the homography and visibility that links camera-object relative motions with occlusions. Third, we propose a confidence module that actively monitors the estimation quality from the pixel correlation distributions obtained in correlation blocks. All these modules are plugged into a Lucas-Kanade (LK) tracking pipeline to obtain both accurate and robust planar object tracking. Our approach outperforms the state-of-the-art methods on public POT and TMT datasets. Its superior performance is also verified on a real-world application, synthesizing high-quality in-video advertisements.","PeriodicalId":72676,"journal":{"name":"Computer vision - ECCV ... : ... European Conference on Computer Vision : proceedings. European Conference on Computer Vision","volume":"26 1","pages":"701-718"},"PeriodicalIF":0.0,"publicationDate":"2022-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77366297","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-09-19DOI: 10.48550/arXiv.2209.08790
Jiefeng Li, Siyuan Bian, Chaoshun Xu, Gang Liu, Gang Yu, Cewu Lu
3D human pose estimation from a monocular video has recently seen significant improvements. However, most state-of-the-art methods are kinematics-based, which are prone to physically implausible motions with pronounced artifacts. Current dynamics-based methods can predict physically plausible motion but are restricted to simple scenarios with static camera view. In this work, we present D&D (Learning Human Dynamics from Dynamic Camera), which leverages the laws of physics to reconstruct 3D human motion from the in-the-wild videos with a moving camera. D&D introduces inertial force control (IFC) to explain the 3D human motion in the non-inertial local frame by considering the inertial forces of the dynamic camera. To learn the ground contact with limited annotations, we develop probabilistic contact torque (PCT), which is computed by differentiable sampling from contact probabilities and used to generate motions. The contact state can be weakly supervised by encouraging the model to generate correct motions. Furthermore, we propose an attentive PD controller that adjusts target pose states using temporal information to obtain smooth and accurate pose control. Our approach is entirely neural-based and runs without offline optimization or simulation in physics engines. Experiments on large-scale 3D human motion benchmarks demonstrate the effectiveness of D&D, where we exhibit superior performance against both state-of-the-art kinematics-based and dynamics-based methods. Code is available at https://github.com/Jeffsjtu/DnD
{"title":"D&D: Learning Human Dynamics from Dynamic Camera","authors":"Jiefeng Li, Siyuan Bian, Chaoshun Xu, Gang Liu, Gang Yu, Cewu Lu","doi":"10.48550/arXiv.2209.08790","DOIUrl":"https://doi.org/10.48550/arXiv.2209.08790","url":null,"abstract":"3D human pose estimation from a monocular video has recently seen significant improvements. However, most state-of-the-art methods are kinematics-based, which are prone to physically implausible motions with pronounced artifacts. Current dynamics-based methods can predict physically plausible motion but are restricted to simple scenarios with static camera view. In this work, we present D&D (Learning Human Dynamics from Dynamic Camera), which leverages the laws of physics to reconstruct 3D human motion from the in-the-wild videos with a moving camera. D&D introduces inertial force control (IFC) to explain the 3D human motion in the non-inertial local frame by considering the inertial forces of the dynamic camera. To learn the ground contact with limited annotations, we develop probabilistic contact torque (PCT), which is computed by differentiable sampling from contact probabilities and used to generate motions. The contact state can be weakly supervised by encouraging the model to generate correct motions. Furthermore, we propose an attentive PD controller that adjusts target pose states using temporal information to obtain smooth and accurate pose control. Our approach is entirely neural-based and runs without offline optimization or simulation in physics engines. Experiments on large-scale 3D human motion benchmarks demonstrate the effectiveness of D&D, where we exhibit superior performance against both state-of-the-art kinematics-based and dynamics-based methods. Code is available at https://github.com/Jeffsjtu/DnD","PeriodicalId":72676,"journal":{"name":"Computer vision - ECCV ... : ... European Conference on Computer Vision : proceedings. European Conference on Computer Vision","volume":"19 1","pages":"479-496"},"PeriodicalIF":0.0,"publicationDate":"2022-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88062279","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}