Monocular 3D object detection has attracted widespread attention due to its potential to accurately obtain object 3D localization from a single image at a low cost. Depth estimation is an essential but challenging subtask of monocular 3D object detection due to the ill-posedness of 2D to 3D mapping. Many methods explore multiple local depth clues such as object heights and keypoints and then formulate the object depth estimation as an ensemble of multiple depth predictions to mitigate the insufficiency of single-depth information. However, the errors of existing multiple depths tend to have the same sign, which hinders them from neutralizing each other and limits the overall accuracy of combined depth. To alleviate this problem, we propose to increase the complementarity of depths with two novel designs. First, we add a new depth prediction branch named complementary depth that utilizes global and efficient depth clues from the entire image rather than the local clues to reduce the similarity of depth predictions. Second, we propose to fully exploit the geometric relations between multiple depth clues to achieve complementarity in form. Benefiting from these designs, our method achieves higher complementarity. Experiments on the KITTI bench-mark demonstrate that our method achieves state-of-the-art performance without introducing extra data. In addition, complementary depth can also be a lightweight and plug-and-play module to boost multiple existing monocular 3d object detectors. Code is available at https://github.com/elvintanhust/MonoCD.

Recently, few-shot learning (FSL) has received increasing attention because of difficulties in sample collection in some application scenarios, such as maritime surveillance using synthetic aperture radar (SAR) or infrared images. In real situations of such scenarios, it is a common requirement that the model can recognize novel classes incrementally, namely class-incremental learning (CIL). Considering the above requirement, a novel problem that recognizes novel classes incrementally when both the base and novel class samples are scarce is proposed in this article. It is called complete few-shot CIL (C-FSCIL) for distinguishing from the FSCIL that assumes sufficient samples of base classes. Specifically, the following challenges of C-FSCIL are focused on: 1) distance measurement is used for recognizing novel classes incrementally, but the encoder is difficult to be learned well when base class samples are scarce, making some features unsuitable for calculating the distance, decreasing the performance and 2) the catastrophic forgetting problem becomes more difficult to be alleviated than that in FSCIL because of the scarcity of base class samples. To tackle both challenges, mine-distill-prototypes (MDP) algorithm is proposed, which consists of two parts: 1) prototypes-distillation (PD) network is proposed to learn to distill the features and prototypes into a lower dimensional in which ineffective features are eliminated and 2) the prototypes-weight (PW) network and the prototypes-selection (PS) training strategy are proposed for the catastrophic forgetting problem, which aims to capture the relationship between the base and novel prototypes. The superior performance of the proposed algorithm is demonstrated by the experiments on three datasets.

Estimating keypoint scale and orientation is crucial to extracting invariant features under significant geometric changes. Recently, the estimators based on self-supervised learning have been designed to adapt to complex imaging conditions. Such learning-based estimators generally predict a single scalar for the keypoint scale or orientation, called hard estimators. However, hard estimators are difficult to handle the local patches containing structures of different objects or multiple edges. In this paper, a Soft Self-Supervised Estimator (S3Esti) is proposed to overcome this problem by learning to predict multiple scales and orientations. S3Esti involves three core factors. First, the estimator is constructed to predict the discrete distributions of scales and orientations. The elements with high confidence will be kept as the final scales and orientations. Second, a probabilistic covariant loss is proposed to improve the consistency of the scale and orientation distributions under different transformations. Third, an optimization algorithm is designed to minimize the loss function, whose convergence is proved in theory. When combined with different keypoint extraction models, S3Esti generally improves over 50% accuracy in image matching tasks under significant viewpoint changes. In the 3D reconstruction task, S3Esti decreases more than 10% reprojection error and improves the number of registered images. [code release]

Human beings acquire the ability of image classification through visual concept learning, in which the process of concept formation involves intertwined searches of common properties and concept descriptions. However, in most image classification algorithms using deep convolutional neural network (ConvNet), the representation space is constructed under the premise that concept descriptions are fixed as one-hot codes, which limits the mining of properties and the ability of identifying unseen samples. Inspired by this, we propose a learning strategy of visual concept formation (LSOVCF) based on the ConvNet, in which the two intertwined parts of concept formation, i.e. feature extraction and concept description, are learned together. First, LSOVCF takes sample response in the last layer of ConvNet to induct concept description being assumed as Gaussian distribution, which is part of the training process. Second, the exploration and experience loss is designed for optimization, which adopts experience cache pool to speed up convergence. Experiments show that LSOVCF improves the ability of identifying unseen samples on cifar10, STL10, flower17 and ImageNet based on several backbones, from the classic VGG to the SOTA Ghostnet. The code is available at \url{https://github.com/elvintanhust/LSOVCF}.

Built-up area is one of the most important objects in remote sensing image analysis, therefore extracting built-up area automatically has attracted wide attention. Deep convolution neural network (CNN) was proposed to improve poor generalization ability of artificial features which had been adopted by traditional automatic extraction methods. In this paper, a more efficient CNN model is proposed to extract the deep features of remote sensing images, and then a graph model based on deep features is constructed to the full image for built-up area extraction. The experiments demonstrate that it has very good performance on the satellite remote sensing image data set.

Built-up area has been one of the most important objects to be extracted in remote sensing images. Several factors such as complex structure, diverse texture and varied background, bring the challenges for the task of built-up area extraction. In this paper, a multiple input structure of deep convolution neural network (CNN) is proposed to extract built-up area automatically, which can fuse the information of panchromatic and multispectral remote sensing image. The image patch based classification results are further refined by postprocessing of segmentation techniques. The experiments demonstrate that the proposed method has better generalization ability compared to the state-of-the-art method, and the overall classification accuracy is above 98%.

Extraction of objects such as buildings from very high resolution(VHR) remote sensing imagery is an important task nowadays. In practical application, the extraction precision is normally satisfied through interactive manual input. Therefore, we propose a semi-automatic building extraction framework with an energy minimization model, which includes two stages: the first stage generates the coarse information of foreground and background, and the second stage extracts the building object by applying energy minimization model. More specifically, in the first stage, the VHR imagery is grouped into small superpixels which are roughly merged into background or foreground according to a line drawn manually. Based on the coarse information of the first stage, we apply the energy minimization model to obtain the precise building object using graph cut optimization. Experimental results indicate our building extraction framework can precisely extract the buildings with different shapes.