Sync with VCLab website + arXiv index trigger

Rebuild org README with website branding (logo, brand color, six directions, PQ816, OPPO), and embed all arXiv URLs to trigger HF Papers indexing.

README.md

@@ -7,77 +7,22 @@ sdk: static
 pinned: false
 ---
 
-# PolyU VCLab
+# PolyU VCLab -- Visual Computing Lab
 
-## Visual Computing Lab, The Hong Kong Polytechnic University
-
-### Official Website -> [polyu-vclab.github.io](https://polyu-vclab.github.io/)
-
-Directed by [**Chair Professor Lei Zhang**](https://www4.comp.polyu.edu.hk/~cslzhang/) (IEEE Fellow) | In long-term collaboration with **OPPO Research Institute**.
+**Visual Computing Lab** at **The Hong Kong Polytechnic University**, directed by [**Chair Professor Lei Zhang**](https://www4.comp.polyu.edu.hk/~cslzhang/) (IEEE Fellow), in long-term collaboration with **OPPO Research Institute**.
 
 > *"Y learning and beyond -- for future visual enhancement and understanding."*
 
-**Quick links:** [Official Website](https://polyu-vclab.github.io/) | [Prof. Lei Zhang](https://www4.comp.polyu.edu.hk/~cslzhang/) | [Google Scholar](https://scholar.google.com/citations?user=tAK5l1IAAAAJ) | [GitHub](https://github.com/PolyU-VCLab)
-
----
-
-## About the Lab
-
-The **Visual Computing Lab (VCLab)** at **The Hong Kong Polytechnic University** works across the full stack of modern visual intelligence: from low-level image and video restoration, through multimodal perception and reasoning, to generative models, 3D reconstruction, new network architectures, and rigorous evaluation benchmarks.
-
-Our work regularly appears at **CVPR, ICCV, ECCV, NeurIPS, ICLR, TPAMI and IJCV**. Many of our methods have been deployed on **hundreds of millions of mobile devices** through our long-term collaboration with **OPPO Research Institute**.
-
-**Office:** PQ816, Department of Computing, PolyU, Hung Hom, Kowloon, Hong Kong
-**Contact:** `cslzhang [at] comp.polyu.edu.hk`
-
----
-
-## Research Directions
-
-Our work is organized into different Collections on this page. Each Collection is kept up-to-date with the latest papers, code, and models -- just click to explore.
-
-**01 - Image / Video Restoration, Enhancement & Quality Assessment**
-Real-world super-resolution, denoising, deblurring, HDR, video restoration, perceptual IQA.
-[Open Collection >](https://huggingface.co/collections/VCLab-HKPU/image-video-restoration-enhancement-and-quality-assessment-69e88641cae9ce5e7cd7d102)
+### Links
 
-**02 - Multimodal Perception, Understanding & Reasoning**
-MLLM-driven visual perception, grounding and reasoning.
-[Open Collection >](https://huggingface.co/collections/VCLab-HKPU/multimodal-perception-understanding-and-reasoning-69e88630a9632f74ae6235a9)
+[Official Website](https://polyu-vclab.github.io/)  |  [GitHub](https://github.com/PolyU-VCLab)  |  [Director's Homepage](https://www4.comp.polyu.edu.hk/~cslzhang/)  |  [Google Scholar](https://scholar.google.com/citations?user=tAK5l1IAAAAJ)
 
-**03 - Image & Video Synthesis and Generation**
-Accelerating, distilling and improving diffusion / AR / DiT generative models.
-[Open Collection >](https://huggingface.co/collections/VCLab-HKPU/image-and-video-synthesis-and-generation-69e88631f170be22735da0ba)
+### Research
 
-**04 - 3D Perception, Reconstruction & Generation**
-Sensing, reconstructing, synthesizing and editing high-fidelity 3D worlds.
-[Open Collection >](https://huggingface.co/collections/VCLab-HKPU/3d-perception-reconstruction-and-generation-69e8863209efbb5392f99651)
+Our work is organized into the six Collections on this page, covering **image/video restoration**, **multimodal understanding**, **generative models**, **3D**, **architectures & training**, and **benchmarks & datasets**.
 
-**05 - Architecture & Training Paradigms**
-New architectures for ViT / LLM / VLM, and efficient, decentralized training.
-[Open Collection >](https://huggingface.co/collections/VCLab-HKPU/architecture-and-training-paradigms-69e886412df265f56289c0e7)
-
-**06 - Benchmarks & Datasets**
-Evaluation benchmarks and training datasets for the visual computing community.
-[Open Collection >](https://huggingface.co/collections/VCLab-HKPU/benchmarks-and-datasets-69e8863311eb2c5cc3ac1d9c)
-
-For the **complete publication list**, please visit Prof. Zhang's [publication page](https://www4.comp.polyu.edu.hk/~cslzhang/papers.htm) and the lab's [official website](https://polyu-vclab.github.io/).
-
----
-
-## Industrial Collaboration
-
-We maintain a long-term strategic collaboration with **OPPO Research Institute**, where many of our algorithms (denoising, super-resolution, HDR, face restoration, low-light enhancement, etc.) have been productized in consumer imaging systems deployed on hundreds of millions of mobile devices.
-
----
-
-## Join Us
-
-We are recruiting **PhD students** (jointly trained with OPPO Research Institute), **postdocs**, and **research interns** across all six research directions.
-
-Send a CV and a brief statement of research interests to **`cslzhang [at] comp.polyu.edu.hk`**, or visit the [Join Us](https://polyu-vclab.github.io/#join-us) page on our website.
-
----
+For the complete publication list, news, team info and open positions, please visit our **[official website](https://polyu-vclab.github.io/)**.
 
-For the most up-to-date information about our lab, people and news, please visit our official website at **[polyu-vclab.github.io](https://polyu-vclab.github.io/)**.
+### Contact
 
-*(c) Visual Computing Lab, The Hong Kong Polytechnic University.*
+PQ816, Department of Computing, PolyU, Hung Hom, Hong Kong  |  `cslzhang [at] comp.polyu.edu.hk`