From the 2 of 16 linked papers with an AI index.
16 papers
RealDenseFace: Real-time Monocular 3D Face Reconstruction from Dense UV-space Priors
Linzhou Li, Tianjia Shao, Kun Zhou
Recent monocular 3D face reconstruction methods achieve high fidelity by fitting a 3D Morphable Model (3DMM) to dense priors predicted by networks, but the optimization stage is co…
GroundShot: Visually Consistent Multi-Shot Long Video Generation via Entity-Grounded Shot Scheduling
Yixuan Lai, Tianjia Shao, Kun Zhou +3
GroundShot is a training-free, model-agnostic framework that improves visual consistency in multi-shot video generation by maintaining an online entity-level visual memory and sche…
Direct Object-Level Reconstruction via Probabilistic Gaussian Splatting
Shuai Guo, Ao Guo, Junchao Zhao +5
The paper introduces a method for efficient single-object 3D reconstruction that integrates foreground‑background probability cues into Gaussian splatting, using probabilistic mask…
Sparse-to-Complete: From Sparse Image Captures to Complete 3D Scenes
Yiyang Shen, Yin Yang, Kun Zhou +1
We introduce S2C-3D, a novel sparse-view 3D reconstruction framework for high-fidelity and complete scene reconstruction from as few as six to eight images. Our framework features…
High-Fidelity Mobile Avatars with Pruned Local Blendshapes
Youyi Zhan, He Wang, Tianjia Shao +1
We propose a method to reconstruct high-fidelity human avatars from multi-view video that can run on mobile devices. Many works can model high-quality Gaussian-based full-body avat…
Slot-ID: Identity-Preserving Video Generation from Reference Videos via Slot-Based Temporal Identity Encoding
Yixuan Lai, He Wang, Kun Zhou +1
Producing prompt-faithful videos that preserve a user-specified identity remains challenging: models need to extrapolate facial dynamics from sparse reference while balancing the t…