Remarkable results have been achieved by DCNN based self-supervised depth estimation approaches. However, most of these approaches can only handle either day-time or night-time images, while their performance degrades...
详细信息
In this paper, we present a large-scale detailed 3D face dataset, FaceScape, and propose a novel algorithm that is able to predict elaborate riggable 3D face models from a single image input. FaceScape dataset provide...
详细信息
Differentiable Architecture Search (DARTS) improves the efficiency of architecture search by learning the architecture and network parameters end-to-end. However, the intrinsic relationship between the architecture’s...
详细信息
ISBN:
(纸本)9781665428132
Differentiable Architecture Search (DARTS) improves the efficiency of architecture search by learning the architecture and network parameters end-to-end. However, the intrinsic relationship between the architecture’s parameters is neglected, leading to a sub-optimal optimization process. The reason lies in the fact that the gradient descent method used in DARTS ignores the coupling relationship of the parameters and therefore degrades the optimization. In this paper, we address this issue by formulating DARTS as a bi-linear optimization problem and introducing an Interactive Differentiable Architecture Search (IDARTS). We first develop a backtracking backpropagation process, which can decouple the relationships of different kinds of parameters and train them in the same framework. The backtracking method coordinates the training of different parameters that fully explore their interaction and optimize training. We present experiments on the CIFAR10 and ImageNet datasets that demonstrate the efficacy of the IDARTS approach by achieving a top-1 accuracy of 76.52% on ImageNet without additional search cost vs. 75.8% with the state-of-the-art PC-DARTS.
The Xiejiawan Formation of early Devonian age in the Longmenshan area of Sichuan Province, China, is a shelf facies that consists of three types of carbonate-siliciclastic deposits: mixed near-shore, clastic mixed she...
详细信息
Out-of-town recommendation is designed for those users who leave their home-town areas and visit the areas they have never been to before. It is challenging to recommend Point-of-Interests (POIs) for out-of-town users...
详细信息
Facial age estimation is an important yet very challenging problem in computer vision. To improve the performance of facial age estimation, we first formulate a simple standard baseline and build a much strong one by ...
详细信息
Existing spatial object recommendation algorithms generally treat objects identically when ranking them. However, spatial objects often cover different levels of spatial granularity and thereby are heterogeneous. For ...
详细信息
Based on the global typical karst characteristics reported in recent years and the latest oil and gas exploration results, it is found that the characteristics of the karst reservoirs in the Ordos Basin are significan...
详细信息
Based on the global typical karst characteristics reported in recent years and the latest oil and gas exploration results, it is found that the characteristics of the karst reservoirs in the Ordos Basin are significantly different from those of traditional karst reservoirs, indicating that the main contribution to the formation of the karst reservoirs in the region may not have been karst dissolution. The previous misunderstanding of the formation mechanism of karst reservoirs suggests that we should think about the formation and development processes of Ordovician karst and the reservoir pores in the Ordos Basin from another perspective. In this study, a combination of karstology, geography, sedimentology, and reservoir geology was used to investigate the formation of reservoirs. Based on the original understanding of the control of large-scale landforms on the sediments in sedimentary basins, this study focuses more on the dynamic spatial-temporal relationships between the evolutions of the landforms, karst, and reservoirs. By studying the petrological characteristics, pore filling, karst zonation, and planation model of the weathering crust reservoir in the Ordos Basin, it is determined that the main reservoir space was formed in advance rather than in the supergene dissolution period. A gypsum-bearing dolomite flat was the material basis for reservoir development and later karst planation transformation. The penecontemporaneous dissolution in the Ordos Basin dominated the spatial development of the early dolomite reservoirs. The traditional karst cycle is actually caused by the erosion of the bottom of the karst floor on the double-layer leveling surface. The karst planation in the Caledonian played a discontinuous and destructive filling role throughout the entire reservoir zone by changing the diagenetic environment, the main contribution of which was communication between the pores. Therefore, the weathered crust reservoir in the Ordos Basin is a residual res
deeplearning has achieved a great success in face recognition (FR), however, few existing models take hierarchical multi-scale local features into consideration. In this work, we propose a hierarchical pyramid divers...
详细信息
ISBN:
(数字)9781728171685
ISBN:
(纸本)9781728171692
deeplearning has achieved a great success in face recognition (FR), however, few existing models take hierarchical multi-scale local features into consideration. In this work, we propose a hierarchical pyramid diverse attention (HPDA) network. First, it is observed that local patches would play important roles in FR when the global face appearance changes dramatically. Some recent works apply attention modules to locate local patches automatically without relying on face landmarks. Unfortunately, without considering diversity, some learned attentions tend to have redundant responses around some similar local patches, while neglecting other potential discriminative facial parts. Meanwhile, local patches may appear at different scales due to pose variations or large expression changes. To alleviate these challenges, we propose a pyramid diverse attention (PDA) to learn multi-scale diverse local representations automatically and adaptively. More specifically, a pyramid attention is developed to capture multi-scale features. Meanwhile, a diverse learning is developed to encourage models to focus on different local patches and generate diverse local features. Second, almost all existing models focus on extracting features from the last convolutional layer, lacking of local details or small-scale face parts in lower layers. Instead of simple concatenation or addition, we propose to use a hierarchical bilinear pooling (HBP) to fuse information from multiple layers effectively. Thus, the HPDA is developed by integrating the PDA into the HBP. Experimental results on several datasets show the effectiveness of the HPDA, compared to the state-of-the-art methods.
Recently, context reasoning using image regions beyond local convolution has shown great potential for scene parsing. In this work, we explore how to incorperate the linguistic knowledge to promote context reasoning o...
详细信息
暂无评论