The intricate task of precisely segmenting retinal vessels from images, which is critical for diagnosing various eye diseases, presents significant challenges for models due to factors such as scale variation, complex anatomical patterns, low contrast, and limitations in training data. Building on these challenges, we offer novel contributions spanning model architecture, loss function design, robustness, and real-time efficacy. To comprehensively address these challenges, a new U-Net-like, lightweight Transformer network for retinal vessel segmentation is presented. By integrating MobileViT+ and a novel local representation in the encoder, our design emphasizes lightweight processing while capturing intricate image structures, enhancing vessel edge precision. A novel joint loss is designed, leveraging the characteristics of weighted cross-entropy and Dice loss to effectively guide the model through the task's challenges, such as foreground-background imbalance and intricate vascular structures. Exhaustive experiments were performed on three prominent retinal image databases. The results underscore the robustness and generalizability of the proposed LiViT-Net, which outperforms other methods in complex scenarios, especially in intricate environments with fine vessels or vessel edges. Importantly, optimized for efficiency, LiViT-Net excels on devices with constrained computational power, as evidenced by its fast performance. To demonstrate the model proposed in this study, a freely accessible and interactive website was established (https://hz-t3.matpool.com:28765?token=aQjYR4hqMI), revealing real-time performance with no login requirements.

As an interesting and challenging problem, generating image caption automatically has attracted increasingly attention in natural language processing and computer vision communities. In this paper, we propose an end-to-end deep learning approach for image caption generation. We leverage image feature information at specific location every moment and generate the corresponding caption description through a semantic attention model. The end-to-end framework allows us to introduce an independent recurrent structure as an attention module, derived by calculating the similarity between image feature sequence and semantic word sequence. Additionally, our model is designed to transfer the knowledge representation obtained from the English portion into the Chinese portion to achieve the cross-lingual image captioning. We evaluate the proposed model on the most popular benchmark datasets. We report an improvement of 3.9% over existing state-of-the-art approaches for cross-lingual image captioning on the Flickr8k CN dataset on CIDEr metric. The experimental results demonstrate the effectiveness of our attention model.

Despite recent successes in hand pose estimation from RGB images or depth maps, inherent challenges remain. RGB-based methods suffer from heavy self-occlusions and depth ambiguity. Depth sensors rely heavily on distance and can only be used indoors, thus there are many limitations to the practical application of depth-based methods. The aforementioned challenges have inspired us to combine the two modalities to offset the shortcomings of the other. In this paper, we propose a novel RGB and depth information fusion network to improve the accuracy of 3D hand pose estimation, which is called CrossFuNet. Specifically, the RGB image and the paired depth map are input into two different subnetworks, respectively. The feature maps are fused in the fusion module in which we propose a completely new approach to combine the information from the two modalities. Then, the common method is used to regress the 3D key-points by heatmaps. We validate our model on two public datasets and the results reveal that our model outperforms the state-of-the-art methods.

At present, the rate control algorithm for multiview high-efficiency video coding (MV-HEVC) does not have the capability of efficient coding tree unit(CTU) layer bit allocation, and the video quality varies greatly for sequences with sudden scene changes or large motions. To overcome this limitation, this paper proposes a rate control algorithm for MV-HEVC based on scene detection. Firstly, we established ρ domain rate control model based on multi-objective optimization. Then, it uses image similarity to make reasonable bit allocation among viewpoints. If the video scene is switched, the image similarity is recalculated, and then the correlation between the weights of the interview point rates and the correlation between the viewpoints are analyzed. Finally, the frame layer rate control considers the layer B-frame and other factors in allocating the code rate, and the basic unit layer rate control adopts different quantization methods according to the content complexity of the CTU. Experimental results show that the proposed rate control algorithm can maintain good coding efficiency and decrease the average video quality variation by 25.29%.

With the rapid growth of population and vehicles, issues such as traffic congestion are becoming increasingly apparent. Parking guidance and information (PGI) systems are becoming more critical, with one of the most important tasks being the prediction of traffic flow in parking lots. Predicting parking traffic can effectively improve parking efficiency and alleviate traffic congestion, traffic accidents, and other problems. However, due to the complex characteristics of parking spatio-temporal data, high levels of noise, and the intricate influence of external factors, there are three challenges to predicting parking traffic in a city effectively: (1) how to better model the nonlinear, asymmetric, and complex spatial relationships among parking lots; (2) how to model the temporal autocorrelation of parking flow more accurately for each parking lot, whether periodic or aperiodic; and (3) how to model the correlation between external influences, such as holiday weekends, POIs (points of interest), and weather factors. In this context, this paper proposes a parking lot traffic prediction model based on the fusion of multifaceted spatio-temporal features (MFF-STGCN). The model consists of a feature embedding module, a spatio-temporal attention mechanism module, and a spatio-temporal convolution module. The feature embedding module embeds external features such as weekend holidays, geographic POIs, and weather features into the time series, the spatio-temporal attention mechanism module captures the dynamic spatio-temporal correlation of parking traffic, and the spatio-temporal convolution module captures the spatio-temporal features by using graph convolution and gated recursion units. Finally, the outputs of adjacent time series, daily series, and weekly series are weighted and fused to obtain the final prediction results, thus predicting the parking lot traffic flow more accurately and effectively. Results on real datasets demonstrate that the proposed model enhances prediction performance.