Traditional video frame interpolation methods based on deep convolutional neural networks face challenges in handling large motions. Their performance is limited by the fact that convolutional operations cannot directly integrate the rich temporal and spatial information of inter-frame pixels, and these methods rely heavily on additional inputs such as optical flow to model motion. To address this issue, we develop a novel framework for video frame interpolation that uses Transformer to efficiently model the long-range similarity of inter-frame pixels. Furthermore, to effectively aggregate spatio-temporal features, we design a novel attention mechanism divided into temporal attention and spatial attention. Specifically, spatial attention is used to aggregate intra-frame information, integrating both attention and convolution paradigms through the simple mapping approach. Temporal attention is used to model the similarity of pixels on the timeline. This design achieves parallel processing of these two types of information without extra computational cost, aggregating information in the space–time dimension. In addition, we introduce a context extraction network and multi-scale prediction frame synthesis network to further optimize the performance of the Transformer. Our method and state-of-the-art methods are extensively quantitatively and qualitatively experimented on various benchmark datasets. On the Vimeo90K and UCF101 datasets, our model achieves improvements of 0.09 dB and 0.01 dB in the PSNR metrics over UPR-Net-large, respectively. On the Vimeo90K dataset, our model outperforms FLAVR by 0.07 dB, with only 40.56% of its parameters. The qualitative results show that for complex and large-motion scenes, our method generates sharper and more realistic edges and details.

Image colorization, a pivotal aspect of computer vision, employs advanced algorithms to transform grayscale images into realistic colors. This task is inherently challenging due to the need to balance colorfulness and fidelity while preserving local spatial structures and eliminating ghosting effect. To address these issues, Our research introduces a novel pipeline leveraging Stable Diffusion for image colorization, guided by color hint points or textual descriptions. Compared to current text-to-image model, Our key contributions include multi-modal input flexibility, a trainable pixel-level encoder and a controllable feature modulation block. The multi-modal input flexibility allows for the simultaneous use of grayscale images with color hint points and textual descriptions, facilitating the generation of colorized outputs with greater precision and alignment with user instructions. The trainable pixel-level encoder extracts multi-scale features from input images, guiding the diffusion process to capture generative diffusion prior for image colorization, thereby achieving better consistency between the input and output images. Additionally, the controllable feature modulation block is introduced to strike a balance between colorfulness and precision through an adjustable coefficient α. By integrating Stable Diffusion with these innovative guidance advancements, our model overcomes previous limitations and showcases the potential of advanced generative models to produce highly realistic and contextually appropriate colorized images, significantly impacting applications such as historical restoration and contemporary creative processes.

Dynamic multi-exposure high dynamic range (HDR) image reconstruction remains challenging because it must preserve globally consistent luminance and structure while recovering fine-grained local textures from low dynamic range (LDR) inputs corrupted by saturation, under-exposure, and motion-induced artifacts. Existing CNN-based methods are effective at local detail restoration but remain limited in global context modeling, whereas Transformer-based methods improve long-range interaction but can still weaken local-detail refinement. Current hybrid designs suggest that the two representation types are complementary, but they do not fully address branch specialization, cross-branch collaboration, and local-feature reliability control. To address this gap, we propose a dual-branch Transformer-CNN framework with a base branch built on Window-based Residual Transformer Blocks (WRTBs), a detail branch equipped with Detail-Aware Gating (DAG) for reliability-aware local refinement, and Bidirectional Cross-Branch Fusion (BCBF) for stage-wise collaboration between the two branches. Experiments on Kalantari17, the Tel benchmark, and Challenge123 show that the proposed design remains competitive on the standard benchmark, achieving the best HDR-VDP2 and tied best with $\mu$-SSIM on Kalantari17, while yielding clearer gains on the more challenging Tel and Challenge123 benchmarks.