Sicong Li
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Sicong Li has a total of 10 co-authors. Here are the top co-authors by number of shared publications:
- Kaice Jiang
Key Laboratory of Opto-Electronic Information Processing, Chinese Academy of Sciences, Shenyang, China
2 shared publications - Feng Zhu
Key Laboratory of Opto-Electronic Information Processing, Chinese Academy of Sciences, Shenyang, China
1 shared publication - Haitao Xu
Shenyang Institute of Automation, Chinese Academy of Sciences
1 shared publication Jianxin Cai
Shenyang Institute of Automation, Chinese Academy of Sciences
1 shared publication- Jianxin Cai
Key Laboratory of Opto-Electronic Information Processing, Chinese Academy of Sciences, Shenyang, China
1 shared publication - Jianyu Wang
Faculty of Robot Science and Engineering, Northeastern University, Shenyang 110169, China
1 shared publication
Latest Publications
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Consumer Electronics
- Vol. 72 (4) , 9872-9884
GC-DETR: Geometry–Color Decoupled Transformer for 3-D Indoor Object Detection in Smart Consumer Devices
3D indoor object detection from point clouds is a core enabling technology for perceptive consumer systems such as service robots, smart home devices, and AR/VR terminals. Although geometric (XYZ) and color (RGB) information are both crucial for reliable on-device scene understanding, most existing approaches simply fuse them in a unified feature space, ignoring their heterogeneous characteristics and causing feature interference. We propose GC-DETR, a Geometry–Color Decoupled and Fused DEtection TRansformer designed to address modal interference in complex indoor environments. GC-DETR follows a decouple–then–fuse paradigm with three key components: a Geometry–Color Disentanglement Module (GCDM), a Dual-branch Feature Pyramid Network (DB-FPN), and a Geometry-guided Cross-Attention Module (G-CAM). Extensive experiments on ScanNetV2 show that GC-DETR achieves state-of-the-art performance, with 78.1% AP@0.25 and 65.1% AP@0.5. Cross-dataset evaluation on SUN RGB-D further demonstrates strong generalization and robustness. Additional ablation and edge-device deployment analyses further support the effectiveness and practical relevance of the proposed design. These results indicate that GC-DETR provides a robust and accurate solution for intelligent perception in next-generation consumer applications.
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Consumer Electronics
- Vol. 72 (4) , 9872-9884
GC-DETR: Geometry–Color Decoupled Transformer for 3-D Indoor Object Detection in Smart Consumer Devices
3D indoor object detection from point clouds is a core enabling technology for perceptive consumer systems such as service robots, smart home devices, and AR/VR terminals. Although geometric (XYZ) and color (RGB) information are both crucial for reliable on-device scene understanding, most existing approaches simply fuse them in a unified feature space, ignoring their heterogeneous characteristics and causing feature interference. We propose GC-DETR, a Geometry–Color Decoupled and Fused DEtection TRansformer designed to address modal interference in complex indoor environments. GC-DETR follows a decouple–then–fuse paradigm with three key components: a Geometry–Color Disentanglement Module (GCDM), a Dual-branch Feature Pyramid Network (DB-FPN), and a Geometry-guided Cross-Attention Module (G-CAM). Extensive experiments on ScanNetV2 show that GC-DETR achieves state-of-the-art performance, with 78.1% AP@0.25 and 65.1% AP@0.5. Cross-dataset evaluation on SUN RGB-D further demonstrates strong generalization and robustness. Additional ablation and edge-device deployment analyses further support the effectiveness and practical relevance of the proposed design. These results indicate that GC-DETR provides a robust and accurate solution for intelligent perception in next-generation consumer applications.
- 27 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 11 (4) , 5166-5173
UNITE-NBV: Uncertainty-Driven and Information-Enhanced Gain Estimation for Next Best View
Next Best View (NBV) algorithms are a critical area of research in 3D reconstruction. They aim to efficiently reconstruct 3D scenes by maximizing information gain from the next optimal viewpoint. However, current NBV methods often neglect the importance of high-quality candidate view sampling, leading to inconsistent quality of the candidate viewpoint set. Moreover, these methods frequently encounter difficulties in extracting effective information for accurate information gain estimation, especially when reconstructing complex objects or large-scale scenes. To address these challenges, we propose UNITE-NBV. Our method achieves accurate next best view selection across various scenes by effectively uniting the proposed high-quality candidate view sampling strategy and information gain estimation network. Specifically, we introduce Spherical Uncertainty Sampling (SUS), a novel candidate viewpoint sampling method. This method calculates the scene’s uncertainty field and maps it into a spherical sampling space. Within this space, the uncertainty distribution guides the sampling of high-quality candidate viewpoints. Additionally, we propose a Multi-Expert Information Gain Network (MEIGN) that performs disentangled encoding and expert encoder sequence processing on features extracted from the reconstructed scene. These processed features are then dynamically fused using sparse gating and Multi-Head Self-Attention, enabling accurate information gain estimation for candidate views. Extensive experimental results on both the small-object dataset ShapeNet and various large-scale 3D scene datasets demonstrate the effectiveness and superior performance of our proposed method. The code will be released.
- 27 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 11 (4) , 5166-5173
UNITE-NBV: Uncertainty-Driven and Information-Enhanced Gain Estimation for Next Best View
Next Best View (NBV) algorithms are a critical area of research in 3D reconstruction. They aim to efficiently reconstruct 3D scenes by maximizing information gain from the next optimal viewpoint. However, current NBV methods often neglect the importance of high-quality candidate view sampling, leading to inconsistent quality of the candidate viewpoint set. Moreover, these methods frequently encounter difficulties in extracting effective information for accurate information gain estimation, especially when reconstructing complex objects or large-scale scenes. To address these challenges, we propose UNITE-NBV. Our method achieves accurate next best view selection across various scenes by effectively uniting the proposed high-quality candidate view sampling strategy and information gain estimation network. Specifically, we introduce Spherical Uncertainty Sampling (SUS), a novel candidate viewpoint sampling method. This method calculates the scene’s uncertainty field and maps it into a spherical sampling space. Within this space, the uncertainty distribution guides the sampling of high-quality candidate viewpoints. Additionally, we propose a Multi-Expert Information Gain Network (MEIGN) that performs disentangled encoding and expert encoder sequence processing on features extracted from the reconstructed scene. These processed features are then dynamically fused using sparse gating and Multi-Head Self-Attention, enabling accurate information gain estimation for candidate views. Extensive experimental results on both the small-object dataset ShapeNet and various large-scale 3D scene datasets demonstrate the effectiveness and superior performance of our proposed method. The code will be released.