S-Lab for Advanced Intelligence

S-Lab for Advanced Intelligence, established in 2020, is a university laboratory at NTU focusing on research and development of cutting-edge AI technologies in computer vision, natural language processing, reinforcement learning, deep learning, and distributed computing. We aim to create impactful applications spanning various strategic areas in partnership with academic, industry, and government organizations.

Featured Dataverses

In order to use this feature you must have at least one published dataverse.

Publish Dataverse

Are you sure you want to publish your dataverse? Once you do so it must remain published.

Publish Dataverse

This dataverse cannot be published because the dataverse it is in has not been published.

Delete Dataverse

Are you sure you want to delete your dataverse? You cannot undelete this dataverse.

1 to 10 of 45 Results

MatAnyone: Stable Video Matting with Consistent Memory Propagation Mar 10, 2025 Yang, Peiqing; Zhou, Shangchen; Zhao, Jixin; Tao, Qingyi; Loy, Chen Change, 2025, "MatAnyone: Stable Video Matting with Consistent Memory Propagation", https://doi.org/10.21979/N9/EN6LQI, DR-NTU (Data), V1 Auxiliary-free human video matting methods, which rely solely on input frames, often struggle with complex or ambiguous backgrounds. To tackle this, we propose MatAnyone, a practical framework designed for target-assigned video matting. Specifically, building on a memory-based fr...
3DEnhancer: Consistent Multi-View Diffusion for 3D Enhancement Mar 10, 2025 Luo, Yihang; Zhou, Shangchen; Lan, Yushi; Pan, Xingang; Loy, Chen Change, 2025, "3DEnhancer: Consistent Multi-View Diffusion for 3D Enhancement", https://doi.org/10.21979/N9/3ARCLF, DR-NTU (Data), V1 Despite advances in neural rendering, due to the scarcity of high-quality 3D datasets and the inherent limitations of multi-view diffusion models, view synthesis and 3D model generation are restricted to low resolutions with suboptimal multi-view consistency. In this study, we pr...
3DTopia-XL: Scaling High-quality 3D Asset Generation via Primitive Diffusion Mar 10, 2025 Chen, Zhaoxi; Tang, Jiaxiang; Dong, Yuhao; Cao, Ziang; Hong, Fangzhou; Lan, Yushi; Wang,Tengfei; Xie, Haozhe; Wu, Tong; Saito, Shunsuke; Pan, Liang; Lin, Dahua; Liu, Ziwei, 2025, "3DTopia-XL: Scaling High-quality 3D Asset Generation via Primitive Diffusion", https://doi.org/10.21979/N9/VMPHUZ, DR-NTU (Data), V1 The increasing demand for high-quality 3D assets across various industries necessitates efficient and automated 3D content creation. Despite recent advancements in 3D generative models, existing methods still face challenges with optimization speed, geometric fidelity, and the la...
Generative Gaussian Splatting for Unbounded 3D City Generation Mar 10, 2025 Xie, Haozhe; Chen, Zhaoxi; Hong, Fangzhou; Liu, Ziwei, 2025, "Generative Gaussian Splatting for Unbounded 3D City Generation", https://doi.org/10.21979/N9/JMQHVG, DR-NTU (Data), V1 3D city generation with NeRF-based methods shows promising generation results but is computationally inefficient. Recently 3D Gaussian Splatting (3D-GS) has emerged as a highly efficient alternative for object-level 3D generation. However, adapting 3D-GS from finite-scale 3D obje...
GaussianAnything: Interactive Point Cloud Latent Diffusion for 3D Generation Feb 5, 2025 Lan, Yushi; Zhou, Shangchen; Lyu, Zhaoyang; Hong, Fangzhou; Yang, Shuai; Dai, Bo; Pan, Xingang; Loy, Chen Change, 2025, "GaussianAnything: Interactive Point Cloud Latent Diffusion for 3D Generation", https://doi.org/10.21979/N9/ZQ85KI, DR-NTU (Data), V1 While 3D content generation has advanced significantly, existing methods still face challenges with input formats, latent space design, and output representations. This paper introduces a novel 3D generation framework that addresses these challenges, offering scalable, high-quali...
Trajectory attention for fine-grained video motion control Feb 4, 2025 Xiao, Zeqi; Ouyang, Wenqi; Zhou, Yifan; Yang, Shuai; Yang, Lei; Si, Jianlou; Pan, Xingang, 2025, "Trajectory attention for fine-grained video motion control", https://doi.org/10.21979/N9/II0EM4, DR-NTU (Data), V1 Recent advancements in video generation have been greatly driven by video diffusion models, with camera motion control emerging as a crucial challenge in creating view-customized visual content. This paper introduces trajectory attention, a novel approach that performs attention...
Denoising as Adaptation: Noise-Space Domain Adaptation for Image Restoration Feb 4, 2025 Liao, Kang; Yue, Zongsheng; Wang, Zhouxia; Loy, Chen Change, 2025, "Denoising as Adaptation: Noise-Space Domain Adaptation for Image Restoration", https://doi.org/10.21979/N9/DMB2QK, DR-NTU (Data), V1 Although learning-based image restoration methods have made significant progress, they still struggle with limited generalization to real-world scenarios due to the substantial domain gap caused by training on synthetic data. Existing methods address this issue by improving data...
SurMo: Surface-based 4D Motion Modeling for Dynamic Human Rendering (CVPR 2024) Jan 16, 2025 Hu, Tao; Hong, Fangzhou; Liu, Ziwei, 2025, "SurMo: Surface-based 4D Motion Modeling for Dynamic Human Rendering (CVPR 2024)", https://doi.org/10.21979/N9/JDZOJE, DR-NTU (Data), V1 Dynamic human rendering from video sequences has achieved remarkable progress by formulating the rendering as a mapping from static poses to human images. However, existing methods focus on the human appearance reconstruction of every single frame while the temporal motion relati...
FashionEngine: Interactive 3D Human Generation and Editing via Multimodal Controls Jan 16, 2025 Hu, Tao; Hong, Fangzhou; Chen, Zhaoxi; Liu, Ziwei, 2025, "FashionEngine: Interactive 3D Human Generation and Editing via Multimodal Controls", https://doi.org/10.21979/N9/WRPWAN, DR-NTU (Data), V1 We present FashionEngine, an interactive 3D human generation and editing system that creates 3D digital humans via user-friendly multimodal controls such as natural languages, visual perceptions, and hand-drawing sketches. FashionEngine automates the 3D human production with thre...
TimeCMA: Towards LLM-Empowered Multivariate Time Series Forecasting via Cross-Modality Alignment Jan 15, 2025 Liu, Chenxi; Xu, Qianxiong; Miao, Hao; Yang, Sun; Zhang, Lingzheng; Long, Cheng; Li, Ziyue; Zhao, Rui, 2025, "TimeCMA: Towards LLM-Empowered Multivariate Time Series Forecasting via Cross-Modality Alignment", https://doi.org/10.21979/N9/V1XDVB, DR-NTU (Data), V1 Multivariate time series forecasting (MTSF) aims to learn temporal dynamics among variables to forecast future time series. Existing statistical and deep learning-based methods suffer from limited learnable parameters and small-scale training data. Recently, large language models...

MatAnyone: Stable Video Matting with Consistent Memory Propagation

Mar 10, 2025

Yang, Peiqing; Zhou, Shangchen; Zhao, Jixin; Tao, Qingyi; Loy, Chen Change, 2025, "MatAnyone: Stable Video Matting with Consistent Memory Propagation", https://doi.org/10.21979/N9/EN6LQI, DR-NTU (Data), V1

Auxiliary-free human video matting methods, which rely solely on input frames, often struggle with complex or ambiguous backgrounds. To tackle this, we propose MatAnyone, a practical framework designed for target-assigned video matting. Specifically, building on a memory-based fr...

3DEnhancer: Consistent Multi-View Diffusion for 3D Enhancement

Mar 10, 2025

Luo, Yihang; Zhou, Shangchen; Lan, Yushi; Pan, Xingang; Loy, Chen Change, 2025, "3DEnhancer: Consistent Multi-View Diffusion for 3D Enhancement", https://doi.org/10.21979/N9/3ARCLF, DR-NTU (Data), V1

Despite advances in neural rendering, due to the scarcity of high-quality 3D datasets and the inherent limitations of multi-view diffusion models, view synthesis and 3D model generation are restricted to low resolutions with suboptimal multi-view consistency. In this study, we pr...

3DTopia-XL: Scaling High-quality 3D Asset Generation via Primitive Diffusion

Mar 10, 2025

Chen, Zhaoxi; Tang, Jiaxiang; Dong, Yuhao; Cao, Ziang; Hong, Fangzhou; Lan, Yushi; Wang,Tengfei; Xie, Haozhe; Wu, Tong; Saito, Shunsuke; Pan, Liang; Lin, Dahua; Liu, Ziwei, 2025, "3DTopia-XL: Scaling High-quality 3D Asset Generation via Primitive Diffusion", https://doi.org/10.21979/N9/VMPHUZ, DR-NTU (Data), V1

The increasing demand for high-quality 3D assets across various industries necessitates efficient and automated 3D content creation. Despite recent advancements in 3D generative models, existing methods still face challenges with optimization speed, geometric fidelity, and the la...

Generative Gaussian Splatting for Unbounded 3D City Generation

Mar 10, 2025

Xie, Haozhe; Chen, Zhaoxi; Hong, Fangzhou; Liu, Ziwei, 2025, "Generative Gaussian Splatting for Unbounded 3D City Generation", https://doi.org/10.21979/N9/JMQHVG, DR-NTU (Data), V1

3D city generation with NeRF-based methods shows promising generation results but is computationally inefficient. Recently 3D Gaussian Splatting (3D-GS) has emerged as a highly efficient alternative for object-level 3D generation. However, adapting 3D-GS from finite-scale 3D obje...

GaussianAnything: Interactive Point Cloud Latent Diffusion for 3D Generation

Feb 5, 2025

Lan, Yushi; Zhou, Shangchen; Lyu, Zhaoyang; Hong, Fangzhou; Yang, Shuai; Dai, Bo; Pan, Xingang; Loy, Chen Change, 2025, "GaussianAnything: Interactive Point Cloud Latent Diffusion for 3D Generation", https://doi.org/10.21979/N9/ZQ85KI, DR-NTU (Data), V1

While 3D content generation has advanced significantly, existing methods still face challenges with input formats, latent space design, and output representations. This paper introduces a novel 3D generation framework that addresses these challenges, offering scalable, high-quali...

Trajectory attention for fine-grained video motion control

Feb 4, 2025

Xiao, Zeqi; Ouyang, Wenqi; Zhou, Yifan; Yang, Shuai; Yang, Lei; Si, Jianlou; Pan, Xingang, 2025, "Trajectory attention for fine-grained video motion control", https://doi.org/10.21979/N9/II0EM4, DR-NTU (Data), V1

Recent advancements in video generation have been greatly driven by video diffusion models, with camera motion control emerging as a crucial challenge in creating view-customized visual content. This paper introduces trajectory attention, a novel approach that performs attention...

Denoising as Adaptation: Noise-Space Domain Adaptation for Image Restoration

Feb 4, 2025

Liao, Kang; Yue, Zongsheng; Wang, Zhouxia; Loy, Chen Change, 2025, "Denoising as Adaptation: Noise-Space Domain Adaptation for Image Restoration", https://doi.org/10.21979/N9/DMB2QK, DR-NTU (Data), V1

Although learning-based image restoration methods have made significant progress, they still struggle with limited generalization to real-world scenarios due to the substantial domain gap caused by training on synthetic data. Existing methods address this issue by improving data...

SurMo: Surface-based 4D Motion Modeling for Dynamic Human Rendering (CVPR 2024)

Jan 16, 2025

Hu, Tao; Hong, Fangzhou; Liu, Ziwei, 2025, "SurMo: Surface-based 4D Motion Modeling for Dynamic Human Rendering (CVPR 2024)", https://doi.org/10.21979/N9/JDZOJE, DR-NTU (Data), V1

Dynamic human rendering from video sequences has achieved remarkable progress by formulating the rendering as a mapping from static poses to human images. However, existing methods focus on the human appearance reconstruction of every single frame while the temporal motion relati...

FashionEngine: Interactive 3D Human Generation and Editing via Multimodal Controls

Jan 16, 2025

Hu, Tao; Hong, Fangzhou; Chen, Zhaoxi; Liu, Ziwei, 2025, "FashionEngine: Interactive 3D Human Generation and Editing via Multimodal Controls", https://doi.org/10.21979/N9/WRPWAN, DR-NTU (Data), V1

We present FashionEngine, an interactive 3D human generation and editing system that creates 3D digital humans via user-friendly multimodal controls such as natural languages, visual perceptions, and hand-drawing sketches. FashionEngine automates the 3D human production with thre...

TimeCMA: Towards LLM-Empowered Multivariate Time Series Forecasting via Cross-Modality Alignment

Jan 15, 2025

Liu, Chenxi; Xu, Qianxiong; Miao, Hao; Yang, Sun; Zhang, Lingzheng; Long, Cheng; Li, Ziyue; Zhao, Rui, 2025, "TimeCMA: Towards LLM-Empowered Multivariate Time Series Forecasting via Cross-Modality Alignment", https://doi.org/10.21979/N9/V1XDVB, DR-NTU (Data), V1

Multivariate time series forecasting (MTSF) aims to learn temporal dynamics among variables to forecast future time series. Existing statistical and deep learning-based methods suffer from limited learnable parameters and small-scale training data. Recently, large language models...

Add Data

Share Dataverse

Link Dataverse

Reset Modifications