Video Diffusion Model
PulseAugur coverage of Video Diffusion Model — every cluster mentioning Video Diffusion Model across labs, papers, and developer communities, ranked by signal.
2 day(s) with sentiment data
-
EgoControl model generates controllable egocentric videos using 3D body poses
Researchers have developed EgoControl, a novel video diffusion model designed for generating egocentric videos with precise control over body motion. This model is trained on egocentric data and conditions future frame …
-
DreamX repurposes video diffusion models for universal latent action learning
Researchers have developed DreamX, a novel method that repurposes a pre-trained video diffusion model to learn universal latent actions. This approach compresses video frames significantly, creating compact representati…
-
PhysForge framework offers advanced material control in 3D asset generation
Researchers have introduced PhysForge, a novel framework designed for enhanced control over material properties in 3D asset generation. This Video Diffusion Model-based system comprises two modules: PhysFit, which uses …
-
StereoEngine generates synthetic stereo data for robot manipulation
Researchers have developed StereoEngine, a novel system for generating synthetic stereo data crucial for training robotic manipulation models. The system employs two methods: one that warps monocular video into a stereo…
-
New method animates glyphs with modal kinetic typography
Researchers have developed a new method for creating modal kinetic typography, which animates vector glyphs to convey semantic concepts while maintaining legibility. This approach utilizes a finite-element eigenproblem …
-
New PhysPlan framework enhances physical realism in AI video generation
Researchers have developed PhysPlan, a new framework designed to improve the physical realism of videos generated by diffusion models. Unlike previous methods that often produce physically implausible sequences, PhysPla…
-
SpatialCrafter generates explorable 3D scenes from single images
Researchers have introduced SpatialCrafter, a novel framework designed for generating explorable 3D scenes from single images. This method addresses limitations in existing approaches by using a global 3D proxy to ensur…
-
New SNM-VFI framework enhances video frame interpolation using motion guidance
Researchers have introduced SNM-VFI, a novel framework for generating intermediate frames in videos. This method utilizes pre-trained optical flow and video diffusion models to guide the generation process with motion-a…
-
Mirror Learning Framework Enhances Imitation Learning with Third-Person Data
Researchers have introduced a novel framework called "mirror learning" to enhance imitation learning by utilizing third-person observational data. This method composes a learned perspective transformation, powered by a …
-
New method reconstructs articulated objects from static images
Researchers have developed a new method for reconstructing articulated objects from a single, static observation, overcoming the limitation of previous techniques that required explicit motion data. This framework utili…
-
New ViDS system uses 3D face tracking for realistic portrait animation
Researchers have developed ViDS, a novel Video Diffusion Shader that utilizes 3D face tracking to create realistic and identity-preserving portrait animations. The system reconstructs a 3D Morphable Model (3DMM) mesh fr…
-
New AI methods enhance 3D rendering, reconstruction, and animation · 4 sources tracked
Researchers are developing new methods to enhance 3D rendering and reconstruction by integrating diffusion models and advanced optimization techniques. One approach, Feature-Guided Diffusion Evolution (FIDE), uses visua…
-
ProxyPose uses video-to-video translation for 6-DoF pose tracking
Researchers have developed ProxyPose, a novel method for tracking the six-degree-of-freedom (6-DoF) pose of objects and surfaces from monocular video. This approach reframes the problem as a video-to-video translation t…
-
New AI model learns joint distribution of videos and camera trajectories
Researchers have developed a novel Video Diffusion Model called Rays as Pixels, which learns a joint distribution of videos and camera trajectories. This model is the first to predict camera poses and generate videos fr…
-
Diffusion Models for Video Generation
Researchers are exploring advanced diffusion models for video generation, addressing challenges like temporal consistency and data scarcity. New methods focus on improving parameterization, such as the v-prediction tech…