4 results about "Computational photography" patented technology

This application discloses a low-light image compression method and system based on an intelligent diffusion model, relating to the fields of image compression and computational photography. The method includes: mapping a low-light image to a low-dimensional latent space using an encoder to obtain a latent feature representation; performing adaptive quantization on the latent feature representation to obtain a discretized compressed feature representation; using the compressed feature representation as a condition, performing an iterative denoising process through a conditional diffusion model to reconstruct a clear latent feature representation; and converting the reconstructed clear latent feature representation to pixel space using a decoder to generate a reconstructed image. The parameters of the conditional diffusion model and encoder are optimized based on an intelligent gain criterion, which is the ratio of information gain to system complexity. This application achieves efficient compression and high-quality reconstruction of low-light images at extremely low bit rates, effectively suppressing noise and restoring details.

The method of snapshot compressive imaging three-dimensional Gaussian splatting reconstruction belongs to the field of computer vision and computational photography. A reconstruction network based on the hybrid architecture of stochastic gradient Langevin dynamics (SGLD) and explicit three-dimensional Gaussian representation is built. Then, multiple physical perception modules are combined to achieve the deep decoupling and reconstruction of the spatiotemporal information of dynamic scenes. Driven by the embedded decoupling deformation field, the deformation field contains a coarse-grained trajectory prediction branch and a fine-grained integral fitting branch: the coarse-grained branch combines low-frequency time coding with latent embedding features, enabling Gaussian primitives to capture global rigid motion trajectories and serving as a temporal regularizer; the fine-grained branch uses high-frequency time coding to predict non-rigid deformation and physically fits the motion blur stripes generated by temporal integration through anisotropic scale stretching. An adaptive density control strategy compatible with physics is introduced. Using the physical integral imaging simulation module, high-fidelity and multi-view consistent high-speed dynamic three-dimensional scene reconstruction results are obtained.

The application discloses an inverse ISP method based on a spiral diffusion model and camera perception adaptation, and relates to the technical field of computer photography and image signal processing. The method comprises the following steps: S1, constructing an inverse ISP training sample, each training sample comprising an RGB image, a target RAW image corresponding to the RGB image and a camera label; S2, constructing an inverse ISP network based on a spiral diffusion model, introducing a time-varying weight map related to pixel intensity at different time steps in the diffusion process; S3, setting a camera perception low-rank adaptation module in the inverse ISP network, comprising a plurality of low-rank adaptation branches corresponding to different camera labels, and selecting a corresponding low-rank adaptation branch to participate in network calculation according to the input camera label; S4, in the training stage, training the inverse ISP network based on the forward probability distribution of the spiral diffusion model; in the sampling stage, inputting the RGB image and the camera label into the trained inverse ISP network, and obtaining the target RAW image based on the reverse iterative sampling process of the spiral diffusion model.

This invention discloses an inverse ISP image reconstruction method based on a diffusion model, belonging to the fields of computer vision and computational photography. This method uses pure noise as the initial input and introduces the target RGB image as a conditional constraint during the diffusion-based inverse denoising process. By fine-tuning the pre-trained diffusion model using ControlNet, the model learns the mapping relationship between RGB and RAW, thereby gradually guiding the reconstruction of a RAW domain image consistent with the conditional RGB content. To improve the structural consistency, detail fidelity, and color / brightness response of the reconstruction results, this invention incorporates attention-guided strategies (such as spatial attention and cross-feature interactive attention) during the conditional injection process. This fully utilizes the edge, texture, and semantic information in the RGB image, suppressing artifacts and detail loss caused by inversion / reconstruction, and enhancing the ability to restore details in extreme exposure areas such as highlights and shadows.