Yuyang Yin

Absolute testing of rotationally symmetric convex surfaces with twin computer-generated holograms

Wei Liu, Runze Li, Zhiyao Ma, SiYe Zhang et autres

cn (code pays fourni par la source)

All-day non-line-of-sight imaging based on Si-SPAD and phase-congruency-based structured ε-regularization

Yuyang Yin, Yingjie Shi, Chenyang Wu, Taotao Qin et autres

cn (code pays fourni par la source)

Fast non-line-of-sight transient data simulation and an open benchmark dataset

Yingjie Shi, Jinye Miao, Taotao Qin, Fuyao Cai et autres

Non-line-of-sight (NLOS) imaging reconstructs the shape and depth of hidden objects from picosecond-resolved transient signals, offering potential applications in autonomous driving, security, and medical diagnostics. However, current NLOS experiments rely on expensive hardware and complex system alignment, limiting their scalability. The limitations …

cn (code pays fourni par la source)

TiP4GEN: Text to Immersive Panorama 4D Scene Generation

Ke Xing, Dejia Xu, Yuyang Yin, Konstantinos N. Plataniotis et autres

With the rapid advancement and widespread adoption of VR/AR technologies, there is a growing demand for the creation of high-quality, immersive dynamic scenes. However, existing generation works predominantly concentrate on the creation of static scenes or narrow perspective-view dynamic scenes, falling short …

cn, ca, us (code pays fourni par la source)

CLIP-GS: Unifying Vision-Language Representation with 3D Gaussian Splatting

Siyu Jiao, Haoye Dong, Yuyang Yin, Zequn Jie et autres

Recent works in 3D multimodal learning have made remarkable progress. However, typically 3D multimodal models are only capable of handling point clouds. Compared to the emerging 3D representation technique, 3D Gaussian Splatting (3DGS), the spatially sparse point cloud cannot depict the texture …

cn, sg, us (code pays fourni par la source)

PanoWorld-X: Generating Explorable Panoramic Worlds via Sphere-Aware Video Diffusion

Yuyang Yin, Fangfu Liu, Mengyu Wang, Hanwen Liang et autres

Generating a complete and explorable 360-degree visual world enables a wide range of downstream applications. While prior works have advanced the field, they remain constrained by either narrow field-of-view limitations, which hinder the synthesis of continuous and holistic scenes, or insufficient camera …

FastMTP: Accelerating LLM Inference with Enhanced Multi-Token Prediction

Yuxuan Cai, Xiaozhuan Liang, Xinghua Wang, Jin Ma et autres

As large language models (LLMs) become increasingly powerful, the sequential nature of autoregressive generation creates a fundamental throughput bottleneck that limits the practical deployment. While Multi-Token Prediction (MTP) has demonstrated remarkable benefits for model training efficiency and performance, its inherent potential for …

TiP4GEN: Text to Immersive Panorama 4D Scene Generation

Ke Xing, Hanwen Liang, Dejia Xu, Yuyang Yin et autres

With the rapid advancement and widespread adoption of VR/AR technologies, there is a growing demand for the creation of high-quality, immersive dynamic scenes. However, existing generation works predominantly concentrate on the creation of static scenes or narrow perspective-view dynamic scenes, falling short …

Matrix-3D: Omnidirectional Explorable 3D World Generation

Zhongqi Yang, Wenhang Ge, Yuqi Li, Jiaqi Chen et autres

Explorable 3D world generation from a single image or text prompt forms a cornerstone of spatial intelligence. Recent works utilize video model to achieve wide-scope and generalizable 3D world generation. However, existing approaches often suffer from a limited scope in the generated …

Martian World Model: Controllable Video Synthesis with Physically Accurate 3D Reconstructions

Longfei Li, Zhiwen Fan, Wenyan Cong, Xinhang Liu et autres

Synthesizing realistic Martian landscape videos is crucial for mission rehearsal and robotic simulation. However, this task poses unique challenges due to the scarcity of high-quality Martian data and the significant domain gap between Martian and terrestrial imagery. To address these challenges, we …

ChineseHarm-Bench: A Chinese Harmful Content Detection Benchmark

Kangwei Liu, Siyuan Cheng, Bozhong Tian, Xiaozhuan Liang et autres

Large language models (LLMs) have been increasingly applied to automated harmful content detection tasks, assisting moderators in identifying policy violations and improving the overall efficiency and accuracy of content review. However, existing resources for harmful content detection are predominantly focused on English, …

Fast Non-Line-of-Sight Transient Data Simulation and an Open Benchmark Dataset

Yingjie Shi, Jinye Miao, Taotao Qin, Fuyao Cai et autres

Non-Line-of-Sight (NLOS) imaging reconstructs the shape and depth of hidden objects from picosecond-resolved transient signals, offering potential applications in autonomous driving, security, and medical diagnostics. However, current NLOS experiments rely on expensive hardware and complex system alignment, limiting their scalability. This manuscript …