This paper aims to reconstruct the object-side light field from the focal stack focusing on different imaging planes. In the forward problem, the focal stack was modeled as projections of the light field. Based on projection modeling, both the filtered-backprojection(FBP) method and the Landweber iterative scheme of solving the inverse problem regarding light field reconstruction from focal stack were derived by applying the methods of computerized tomography(CT). The experimental results show that the high-precision light field can be reconstructed via FBP and Simultaneous-Algebraic-Reconstruction-Technique(SART) algorithm, and depth and surface of the scene can be reconstructed via the reconstructed light field.

Underwater multi-object tracking (UMOT) is an important technology in marine animal ethology. It is affected by complex factors such as scattering, background interference, and occlusion, which makes it a challenging computer vision task. As a result, the stable continuation of trajectories among different targets has been the key to the tracking performance of UMOT tasks. To solve such challenges, we propose an underwater multi-object tracking algorithm based on memory aggregation (UMOTMA) to effectively associate multiple frames with targets. First, we propose a long short-term memory (LSTM)-based memory aggregation module (LSMAM) to enhance memory utilization between multiple frames. Next, LSMAM embeds LSTM into the transformer structure to save and aggregate features between multiple frames. Then, an underwater image enhancement module M_E is introduced to process the original underwater images, which improves the quality and visibility of the underwater images so that the model can extract better features from the images. Finally, LSMAM and M_E are integrated with a backbone network to implement the entire algorithm framework, which can fully utilize the historical information of the tracked targets. Experiments on the UMOT datasets and the underwater fish school datasets show that UMOTMA generally outperforms existing models and can maintain the stability of the target trajectory while ensuring high-quality detection. The code is available via Github.

Pose estimation plays a crucial role in recognizing and analyzing the postures, actions, and movements of humans and animals using computer vision and machine learning techniques. However, bird pose estimation encounters specific challenges, including bird diversity, posture variation, and the fine granularity of posture. To overcome these challenges, we propose VHR-BirdPose, a method that combines Vision Transformer (ViT) and Deep High-Resolution Network (HRNet) with an attention mechanism. VHR-BirdPose effectively extracts features using Vision Transformer’s self-attention mechanism, which captures global dependencies in the images and allows for better capturing of pose details and changes. The attention mechanism is employed to enhance the focus on bird keypoints, improving the accuracy of pose estimation. By combining HRNet with Vision Transformer, our model can extract multi-scale features while maintaining high-resolution details and incorporating richer semantic information through the attention mechanism. This integration of HRNet and Vision Transformer leverages the advantages of both models, resulting in accurate and robust bird pose estimation. We conducted extensive experiments on the Animal Kingdom dataset to evaluate the performance of VHR-BirdPose. The results demonstrate that our proposed method achieves state-of-the-art performance in bird pose estimation. VHR-BirdPose based on bird images is of great significance for the advancement of bird behaviors, ecological understanding, and the protection of bird populations.

In the light field imaging, the geometrical parameters calibration of the light field camera is the foundation of computational imaging techniques such as the calibration of the light field data, digital refocusing, depth reconstruction, all-in-focus imaging and so on. In this paper, we propose a virtual image points based geometrical parameters calibration for focused light field camera. In the forward model, the mapping from the object points to the virtual image points with respect to the main lens, as well as the mapping from the virtual image points to the image points on the detector with respect to the microlens array are established. While calibrating geometrical parameters, the inverse problem of calculating the virtual image points from the corner on the detector is established and the model of calculating the geometrical parameters of the focused light field camera is established. Levenberg-Marquardt algorithm is used to solve the geometrical parameters iteratively. We use the checkerboard for the calibration experiment and validate the calculation via the reprojection of the checkerboard corners.

Depth reconstruction from the light field, as a depth extracting approach, is a vibrant research field in computational imaging. In this paper, depth reconstruction from the light field was approached as an optimization problem by analyzing the forward and inverse models. The forward and inverse models present the connection between the light field and the depth of the scene. We proposed an iterative method for scene depth reconstruction with fidelity from 4D light field data. The objective function of the optimization problem entails three terms, of which the matching term is used as the fidelity term, while the gradient term and classification term constitute the penalty terms. The iterative method minimized the objective function by correcting mismatching pixels iteratively. In this way, the reconstruction accuracy of the scene depth was significantly improved with high robustness, especially in the weak texture and occlusion regions. The experimental results show that the depth map can be reconstructed with high accuracy by the iterative method, in alignment with high depth resolution in the weak texture and occlusion regions. An accurate 3D surface in the field of view can be achieved using the depth information reconstructed from the 4D light field.