Guosheng Lin

Prioritizing Faithfulness: Efficient Zero-Shot Novel View Synthesis with Adaptive Latent Modulation

In-Context Learning with Unpaired Clips for Instruction-based Video Editing

SAMITE: Position Prompted SAM2 with Calibrated Memory for Visual Object Tracking

Puppeteer: Rig and Animate Your 3D Models

Efficiently Maintaining the Multilingual Capacity of MCLIP in Downstream Cross-Modal Retrieval Tasks

ADHMR: Aligning Diffusion-based Human Mesh Recovery via Direct Preference Optimization

Unlocking the Power of SAM 2 for Few-Shot Segmentation

MeshAnything: Artist-Created Mesh Generation with Autoregressive Transformers

ADAPT: Attentive Self-Distillation and Dual-Decoder Prediction Fusion for Continual Panoptic Segmentation

Text-to-Image Rectified Flow as Plug-and-Play Priors

Robust-PIFu: Robust Pixel-aligned Implicit Function for 3D Human Digitalization from a Single Image

U3D: Unlocking the Video Prior for High Fidelity Sparse Novel View Synthesis and 3D Generation

Sync4D: Video Guided Controllable Dynamics for Physics-Based 4D Generation