SAGE: Scene Graph-Aware Guidance and Execution for Long-Horizon Manipulation Tasks

• we propose SAGE, a novel framework for Scene Graph-Aware Guidance and Execution in Long-Horizon Manipulation Tasks

LLGS: Unsupervised Gaussian Splatting for Image Enhancement and Reconstruction in Pure Dark Environment

• we propose an unsupervised multi-view stereoscopic system based on Gaussian Splatting, called Low-Light Gaussian Splatting.

4DGS point flow Convert 3DGS to a time series format and simultaneously use semantic-driven methods to track objects.

FSAG: Enhancing Human-to-Dexterous-Hand Finger-Specific Affordance Grounding via Diffusion Models

• we propose a data-efficient framework, which is designed to bypass robot grasp data collection by exploiting the rich, object-centric semantic priors latent in pretrained generative diffusion models.

Meta-Scoop: A Coarse-to-Precise Policy Learning Framework for Precision Scooping Across Task Variations

• We have proposed a framework that progresses from rough to precise, enabling precise extraction operations in various task variations.

LTCF-Net: A Transformer-Enhanced Dual-Channel Fourier Framework for Low-Light Image Restoration \

Project

• We introduce LTCF-Net, a novel network architecture designed for enhancing low-light images.

Lighting SLAM Paths in Darkness: A Low-Light Augmented SLAM with Hybrid Window Transformer-Mamba \

Project 📺 Video Youtube

• We propose Low-Light Enhanced SLAM with Hybrid Window Transformer-Mamba (HWTM-SLAM), a novel framework that integrates low-light image enhancement as a fundamental front-end component to address the fragility of traditional SLAM systems in dark environments.