Hanzhang Tu3D vision · Neural rendering
Personal research archiveBeijing, China

Hanzhang Tu

PhD candidate · Tsinghua University

Seeing people in 3D.

I work on 3D reconstruction, neural rendering, and real-time human telepresence. My research turns sparse camera observations into dynamic representations of people, making it possible to see and connect beyond a single viewpoint.

Explore research Get in touch
Google Scholar Curriculum vitae
Spatial study / 01Geometry → presence
An original schematic of spatial structure. Explore its layers and viewpoints; this is an illustration, not a research result.
013D reconstruction
02Neural rendering
03Human telepresence

01 / Selected research

From cameras to presence.

Projects connecting geometric reconstruction, human appearance, and real-time communication.

Tele360 telepresence illustration Illustration of a person captured by four cameras, connected by a red dashed line to three remote viewing scenes.

Tele360: Real-Time Feed-Forward Human Reconstruction from Sparse Unposed Cameras

ACM TOG · SIGGRAPH Asia 2026Accepted

A feed-forward system that reconstructs dynamic people from sparse, unposed RGB cameras. By bringing camera estimation, 3D Gaussian reconstruction, and streaming into one pipeline, Tele360 connects multi-camera capture to free-viewpoint remote viewing. The representation is designed to move from live observations to a renderable scene without per-subject optimization.

Hanzhang Tu, Z. Liao, W. Min, J. Zhang, Y. Liu

Quick look

GBC-Splat: Generalizable Gaussian-Based Clothed Human Digitalization under Sparse RGB Cameras

CVPR 2025

Reconstructing clothed people means recovering both their shape and their appearance. GBC-Splat lifts sparse RGB views into a complete human surface and a Gaussian representation, without a body template or subject-specific optimization. The resulting geometry supports free-viewpoint rendering, including people in loose clothing such as dresses and costumes.

Hanzhang Tu, Zhanfeng Liao, Boyao Zhou, Shunyuan Zheng, Xilong Zhou, Liuxin Zhang, QianYing Wang, Yebin Liu

Quick look Paper Conference

Tele-Aloha: A Low-budget and High-authenticity Telepresence System Using Sparse RGB Cameras

SIGGRAPH 2024

A practical telepresence system built around four RGB cameras, a consumer GPU, and an autostereoscopic display. Tele-Aloha connects capture, reconstruction, and streaming to deliver a view-dependent image of a remote person. The complete system explores how real-time neural rendering can support a sense of co-presence without a wearable headset.

Hanzhang Tu, Ruizhi Shao, Xue Dong, Shunyuan Zheng, Hao Zhang, Lili Chen, Meili Wang, Wenyu Li, Siyan Ma, Shengping Zhang, Boyao Zhou, Yebin Liu

Quick look Paper

02 / Further reading

Other publications.

Full list on Google Scholar

A common thread

Making the distant feel present.

From a few images to a shared view of a person: geometry gives us structure, rendering gives us appearance, and real-time systems bring them into a conversation.

CaptureReconstructConnect

03 / About

A little context.

Beijing, China

I am a PhD candidate in the Department of Automation at Tsinghua University, advised by Prof. Yebin Liu. My work sits at the intersection of computer vision and computer graphics, with a focus on reconstructing people and bringing them into shared digital spaces.

I develop feed-forward reconstruction methods, Gaussian representations, and real-time systems for human telepresence. Before my PhD, I received a BEng in Measurement and Control Technology and Instrumentation from Tsinghua University, with a minor in Computer Science. I expect to complete my PhD in June 2027.

  • 2022–2027

    PhD · Control Science and Engineering

    Tsinghua University · Expected June 2027

  • 2018–2022

    BEng · Measurement and Control Technology and Instrumentation

    Tsinghua University · Minor in Computer Science

04 / Contact

Let’s compare perspectives.

For research conversations and collaborations.

thz22@mails.tsinghua.edu.cn
Google Scholar Curriculum vitae

Tele360: Real-Time Feed-Forward Human Reconstruction from Sparse Unposed Cameras

ACM TOG · SIGGRAPH Asia 2026Accepted

Hanzhang Tu, Z. Liao, W. Min, J. Zhang, Y. Liu

Tele360 telepresence illustration Illustration of a person captured by four cameras, connected by a red dashed line to three remote viewing scenes.
Telepresence illustration

TL;DR

Tele360 reconstructs dynamic humans from four to six sparse, unposed RGB cameras. The feed-forward pipeline jointly estimates camera poses and reconstructs dynamic 3D Gaussians, connecting live capture with free-viewpoint telepresence.

Abstract

Live free-viewpoint visualization of real humans is critical for immersive communication and interactive digital experiences. Existing methods either rely on computationally expensive optimization or require calibrated cameras and low-resolution inputs, making real-time high-resolution deployment impractical. In this work, we present Tele360, the first real-time feed-forward system for dynamic human reconstruction and live free-viewpoint visualization from sparse, unposed RGB streams.

Our system jointly estimates camera poses and reconstructs a dynamic 3D Gaussian representation for each time instance in a single forward pass. To achieve this, we start by designing a lightweight sparsity-aware multi-view transformer backbone that tokenizes foreground human regions while preserving global context through a shared scene token. We then employ a fully transformer-based Gaussian decoder to mitigate convolution-induced over-smoothing while keeping decoding sparse and efficient. In addition, we introduce a hybrid feature pyramid that injects multi-scale appearance cues into geometry prediction. We further introduce a lightweight differentiable Levenberg-Marquardt camera refinement layer to enhance multi-view consistency and geometric alignment.

Moreover, to stabilize learning under sparse, unposed inputs, we transfer multi-view geometry priors from large visual-geometry foundation model via teacher-student distillation. Finally, the predicted gaussian maps are streamed with video codecs to remote devices for interactive free-viewpoint rendering. Together, these designs form a balanced pipeline between efficiency and fidelity. Extensive experiments show that Tele360 achieves state-of-the-art visual quality while supporting 2K input-to-rendering in real time at over 25 FPS on a single consumer GPU across both studio benchmarks and challenging in-the-wild scenarios.

Full project page

GBC-Splat: Generalizable Gaussian-Based Clothed Human Digitalization under Sparse RGB Cameras

CVPR 2025

Hanzhang Tu, Zhanfeng Liao, Boyao Zhou, Shunyuan Zheng, Xilong Zhou, Liuxin Zhang, QianYing Wang, Yebin Liu

Paper Conference
Human reconstruction illustration

TL;DR

GBC-Splat is a generalizable, feed-forward approach to clothed-human digitalization. It reconstructs human geometry and Gaussian appearance directly from sparse RGB input, without subject-specific optimization or an SMPL template.

Abstract

We present an efficient approach for generalizable clothed human digitalization, termed GBC-Splat. Unlike previous methods that necessitate per-subject optimizations or discount watertight geometry, the proposed method is dedicated to reconstructing complete human shapes and Gaussian Splatting via sparse view RGB inputs in a feed-forward manner.

We first extract a fine-grained mesh using a combination of implicit occupancy field regression and explicit disparity estimation between views. The reconstructed high-quality geometry allows us to easily anchor Gaussian primitives to mesh surface according to surface normal and texture, which allows 6-DoF photorealistic novel view synthesis. In addition, we introduce a simple yet effective algorithm to subdivide Gaussian primitives in high-frequency areas to further enhance the visual quality.

Without the assistance of human parametric models, our method can tackle loose garments, such as dresses and costumes. Our method outperforms state-of-the-art methods in terms of novel view synthesis while keeping high efficiency, enabling the potential of deployment in real-time applications.

Full project page

Tele-Aloha: A Low-budget and High-authenticity Telepresence System Using Sparse RGB Cameras

SIGGRAPH 2024

Hanzhang Tu, Ruizhi Shao, Xue Dong, Shunyuan Zheng, Hao Zhang, Lili Chen, Meili Wang, Wenyu Li, Siyan Ma, Shengping Zhang, Boyao Zhou, Yebin Liu

Paper
Telepresence illustration

TL;DR

Tele-Aloha is a bidirectional telepresence system for peer-to-peer communication. Sparse RGB capture, neural rendering, and an autostereoscopic display bring a remote person's upper body into a view-dependent, three-dimensional communication experience.

Abstract

In this paper, we present a low-budget and high-authenticity bidirectional telepresence system, Tele-Aloha, targeting peer-to-peer communication scenarios. Compared to previous systems, Tele-Aloha utilizes only four sparse RGB cameras, one consumer-grade GPU, and one autostereoscopic screen to achieve high-resolution (2048 × 2048), real-time (30 fps), low-latency (less than 150 ms) and robust distant communication.

As the core of Tele-Aloha, we propose an efficient novel view synthesis algorithm for upper-body. Firstly, we design a cascaded disparity estimator for obtaining a robust geometry cue. Additionally a neural rasterizer via Gaussian Splatting is introduced to project latent features onto target view and to decode them into a reduced resolution. Further, given the high-quality captured data, we leverage weighted blending mechanism to refine the decoded image into the final resolution of 2K.

Exploiting world-leading autostereoscopic display and low-latency iris tracking, users are able to experience a strong three-dimensional sense even without any wearable head-mounted display device. Altogether, our telepresence system demonstrates the sense of co-presence in real-life experiments, inspiring the next generation of communication.

Full project page