Light field (LF) technology captures information from multiple directions and angles, enabling precise disparity estimation. Recently, matching cost-based approaches have advanced rapidly and shown satisfactory results. However, these methods typically depend on fixed disparity candidates, leading to inadequate utilization of candidates and making them unsuitable for LF scenes with varying baselines. Multi-direction line structures of epipolar-plane images (EPIs) associate multiple viewpoints, adaptively perceiving disparity ranges and accurately matching features in real scenes. In this paper, we propose an Adaptive EPI-Matching Cost for LF disparity estimation, which is proven to enhance the adaptability across datasets with varying baselines. Our approach calculates pixel-level disparity candidates to keep the predicted distribution near the ground truth and matches line structures to improve accuracy. Then, to enhance robustness during the adaptive process, we introduce an intra-EPI extraction module that dynamically establishes correlations in the local EPI while supplementing spatial information. Finally, we present a network named Adaptive EPI-Matching Cost Network (AEMCNet) for LF disparity estimation. Experimental results demonstrate that AEMCNet achieves state-of-the-art performance and robustness on various LF datasets with different baselines. Specifically, on the Sparse Light Field Dataset, our method reduces the mean square error by 49.6%.

As one of the fundamental technologies for scene understanding, semantic segmentation has been widely explored in the last few years. Light field cameras encode the geometric information by simultaneously recording the spatial information and angular information of light rays, which provides us with a new way to solve this issue. In this paper, we propose a high-quality and challenging urban scene dataset, containing 1074 samples composed of real-world and synthetic light field images as well as pixel-wise annotations for 14 semantic classes. To the best of our knowledge, it is the largest and the most diverse light field dataset for semantic segmentation. We further design two new semantic segmentation baselines tailored for light field and compare them with state-of-the-art RGB, video and RGB-D-based methods using the proposed dataset. The outperforming results of our baselines demonstrate the advantages of the geometric information in light field for this task. We also provide evaluations of super-resolution and depth estimation methods, showing that the proposed dataset presents new challenges and supports detailed comparisons among different methods. We expect this work inspires new research direction and stimulates scientific progress in related fields. The complete dataset is available at https://github.com/HAWKEYE-Group/UrbanLF.

Global context information is particularly important for comprehensive scene understanding. It helps clarify local confusions and smooth predictions to achieve fine-grained and coherent results. However, most existing light field processing methods leverage convolution layers to model spatial and angular information. The limited receptive field restricts them to learn long-range dependency in LF structure. In this paper, we propose a novel network based on deep efficient transformers ( i.e., LF-DET) for LF spatial super-resolution. It develops a spatial-angular separable transformer encoder with two modeling strategies termed as sub-sampling spatial modeling and multi-scale angular modeling for global context interaction. Specifically, the former utilizes a sub-sampling convolution layer to alleviate the problem of huge computational cost when capturing spatial information within each sub-aperture image. In this way, our model can cascade more transformers to continuously enhance feature representation with limited resources. The latter processes multi-scale macro-pixel regions to extract and aggregate angular features focusing on different disparity ranges to well adapt to disparity variations. Besides, we capture strong similarities among surrounding pixels by dynamic positional encodings to fill the gap of transformers that lack of local information interaction. The experimental results on both real-world and synthetic LF datasets confirm our LF-DET achieves a significant performance improvement compared with state-of-the-art methods. Furthermore, our LF-DET shows high robustness to disparity variations through the proposed multi-scale angular modeling.

Recently, tracking-by-detection has become a popular paradigm in Multiple-object tracking (MOT) for its concise pipeline. Many current works first associate the detections to form track proposals and then score proposalns by manual functions to select the best. However, long-term tracking information is lost in this way due to detection failure or heavy occlusion. In this paper, the Extendable Multiple Nodes Tracking framework (EMNT) is introduced to model the association. Instead of detections, EMNT creates four basic types of nodes including correct, false, dummy and termination to generally model the tracking procedure. Further, we propose a General Recurrent Tracking Unit (RTU++) to score track proposals by capturing long-term information. In addition, we present an efficient generation method of simulated tracking data to overcome the dilemma of limited available data in MOT. The experiments show that our methods achieve state-of-the-art performance on MOT17, MOT20 and HiEve benchmarks. Meanwhile, RTU++ can be flexibly plugged into other trackers such as MHT, and bring significant improvements. The additional experiments on MOTS20 and CTMC-v1 also demonstrate the generalization ability of RTU++ trained by simulated data in various scenarios.