Lin Chen

Monocular depth estimation (MDE) is a critical computer vision task that enhances environmental perception in fields such as autonomous driving and robot navigation. In recent years, deep learning-based MDE methods have achieved notable progress in these fields. However, achieving robust monocular depth estimation in low-altitude forest environments remains challenging, particularly in scenes with dense and cluttered foliage, which complicates applications in environmental monitoring, agriculture, and search and rescue operations. This paper presents a comprehensive evaluation of state-of-the-art deep learning-based MDE methods on low-altitude forest datasets. The evaluated models include both self-supervised and supervised approaches, employing different network structures such as convolutional neural networks (CNNs) and Vision Transformers (ViTs). We assessed the generalization of these approaches across diverse low-altitude scenarios, specifically focusing on forested environments. A systematic set of evaluation criteria is employed, comprising traditional image-based global statistical metrics as well as geometry-aware metrics, to provide a more comprehensive evaluation of depth estimation performance. The results indicate that most Transformer-based models, such as DepthAnything and Metric3D, outperform traditional CNN-based models in complex forest environments by capturing detailed tree structures and depth discontinuities. Conversely, CNN-based models like MiDas and Adabins struggle with handling depth discontinuities and complex occlusions, yielding less detailed predictions. On the Mid-Air dataset, the Transformer-based DepthAnything demonstrates a 54.2% improvement in RMSE for the global error metric compared to the CNN-based Adabins. On the LOBDM dataset, the CNN-based MiDas has the depth edge completeness error of 93.361, while the Transformer-based Metric3D demonstrates the significantly lower error of only 5.494. These findings highlight the potential of Transformer-based approaches for monocular depth estimation in low-altitude forest environments, with implications for high-throughput plant phenotyping, environmental monitoring, and other forest-specific applications.

Recent developments in deep learning technology have boosted the performance of dense stereo reconstruction. However, the state-of-the-art deep learning-based stereo matching methods are mainly trained using close-range synthetic images. Consequently, the application of these methods in aerial photogrammetry and remote sensing is currently far from straightforward. In this paper, we propose a new disparity estimation network for stereo matching and investigate its generalization abilities in regard to aerial images. First, we propose an end-to-end deep learning network for stereo matching, regularized by disparity gradients, which includes a residual cost volume and a reconstruction error volume in a refinement module, and multiple losses. In order to investigate the influence of the multiple losses, a comprehensive analysis is presented. Second, based on this network trained with synthetic close-range data, we propose a new pipeline for matching high-resolution aerial imagery. The experimental results show that the proposed network improves the disparity accuracy by up to 40% in terms of errors larger than 1 px compared to results when not including the refinement network, especially in areas containing detailed small objects. In addition, in qualitative and quantitative experiments, we are able to show that our model, pre-trained on a synthetic stereo dataset, achieves very competitive sub-pixel geometric accuracy on aerial images. These results confirm that the domain gap between synthetic close-range and real aerial images can be satisfactorily bridged using the proposed new deep learning method for dense image matching.

We propose a novel framework to learn 3D point cloud semantics from 2D multi-view image observations containing pose error. On the one hand, directly learning from the massive, unstructured and unordered 3D point cloud is computationally and algorithmically more difficult than learning from compactly-organized and context-rich 2D RGB images. On the other hand, both LiDAR point cloud and RGB images are captured in standard automated-driving datasets. This motivates us to conduct a "task transfer" paradigm so that 3D semantic segmentation benefits from aggregating 2D semantic cues, albeit pose noises are contained in 2D image observations. Among all difficulties, pose noise and erroneous prediction from 2D semantic segmentation approaches are the main challenges for the task transfer. To alleviate the influence of those factor, we perceive each 3D point using multi-view images and for each single image a patch observation is associated. Moreover, the semantic labels of a block of neighboring 3D points are predicted simultaneously, enabling us to exploit the point structure prior to further improve the performance. A hierarchical full attention network~(HiFANet) is designed to sequentially aggregates patch, bag-of-frames and inter-point semantic cues, with hierarchical attention mechanism tailored for different level of semantic cues. Also, each preceding attention block largely reduces the feature size before feeding to the next attention block, making our framework slim. Experiment results on Semantic-KITTI show that the proposed framework outperforms existing 3D point cloud based methods significantly, it requires much less training data and exhibits tolerance to pose noise. The code is available at https://github.com/yuhanghe01/HiFANet.

In feature based image matching, distinctive features in images are detected and represented by feature descriptors. Matching is then carried out by assessing the similarity of the descriptors of potentially conjugate points. In this paper, we first shortly discuss the general framework. Then, we review feature detection as well as the determination of affine shape and orientation of local features, before analyzing feature description in more detail. In the feature description review, the general framework of local feature description is presented first. Then, the review discusses the evolution from hand-crafted feature descriptors, e.g. SIFT (Scale Invariant Feature Transform), to machine learning and deep learning based descriptors. The machine learning models, the training loss and the respective training data of learning-based algorithms are looked at in more detail; subsequently the various advantages and challenges of the different approaches are discussed. Finally, we present and assess some current research directions before concluding the paper.

A new community and communication type of social networks - online dating - are gaining momentum. With many people joining in the dating network, users become overwhelmed by choices for an ideal partner. A solution to this problem is providing users with partners recommendation based on their interests and activities. Traditional recommendation methods ignore the users' needs and provide recommendations equally to all users. In this paper, we propose a recommendation approach that employs different recommendation strategies to different groups of members. A segmentation method using the Gaussian Mixture Model (GMM) is proposed to customize users' needs. Then a targeted recommendation strategy is applied to each identified segment. Empirical results show that the proposed approach outperforms several existing recommendation methods.