Yunhao Ge

12

Papers

111

Total Citations

Papers (12)

Describe Anything: Detailed Localized Image and Video Captioning

Visual Fact Checker: Enabling High-Fidelity Detailed Caption Generation

BEHAVIOR Vision Suite: Customizable Dataset Generation via Simulation

DreamDistribution: Learning Prompt Distribution for Diverse In-distribution Generation

ArtiScene: Language-Driven Artistic 3D Scene Generation Through Image Intermediary

Neural-Sim: Learning to Generate Training Data with NeRF

Improving Zero-Shot Generalization and Robustness of Multi-Modal Models

A Peek Into the Reasoning of Neural Networks: Interpreting With Structural Visual Concepts

CLR: Channel-wise Lightweight Reprogramming for Continual Learning

Pose Augmentation: Class-agnostic Object Pose Transformation for Object Recognition

"Contributions of Shape, Texture, and Color in Visual Recognition"

3D Copy-Paste: Physically Plausible Object Insertion for Monocular 3D Detection