Robby Tan
Associate Professor
Leads the Vision and AI Lab on computer vision in adverse outdoor conditions, human-centric analysis, fake image detection and agentic multimodal foundation models.
Robby Tan's lab at NUS works on recognition, detection & segmentation, vision-language & multimodal and multimodal & grounded language. 46 papers since 2023, 39 at top venues, mostly CVPR and AAAI. 9 PhD students. 44 alumni. Applications close .
At a glance
- papers, 2023–2026
- 12/ year
- at T1 venues
- 85%
- PhD students
- 9
- MS students
- 0
- alumni listed
- 44
Papers by yearfrom the lab homepage, 2023–2026
- 13 papers2026 · T1 12 · collecting
- 9 papers2025 · T1 6
- 14 papers2024 · T1 13
- 10 papers2023 · T1 8
T1 venuesOtherVenues: CVPR, AAAI, ACL, ECCV, NeurIPS
Research focuspaper titles that use each topic's words
- Recognition, detection & segmentation12 papers
- Vision-language & multimodal9 papers
- Multimodal & grounded language6 papers
- conditions2 papers
- outdoor0 papers
Paper list
From the lab homepage.
202613 papers
- ECCV (T1)Actor as Its Own Critic: Unifying Region Understanding and Localization via CycleGRPO
- AAAI (T1)Aggregating Diverse Cue Experts for AI-Generated Image Detection
- AAAI Conference on Artificial Intelligence, AAAIBridging Day and Night: Target-Class Hallucination Suppression in Unpaired Image Translation
- ACL (T1)CodeJudgeBench: On the Reliability of LLMs as Code Judges
- CVPR (T1)CrossHOI-Bench: A Unified Benchmark for HOI Evaluation across Vision-Language Models and HOI-Specific Methods
- CVPR (T1)GlowGS: Generative Semantic Feature Learning for 3D Gaussian Splatting in Nighttime Glow Scenes
- CVPR (T1)Mind the Gap: Transferring Labels to Align Object Detection Datasets
- ACL (T1)NaturalSloth: Revisiting Denial-of-Service Attacks on Large Language Models
- ACL (T1)ReMedi: Reasoner for Medical Clinical Prediction
- ECCV (T1)Token-Based Affordance Grounding with Large Vision-Language Models
- ECCV (T1)UC-VLM: Consistency-Driven Learning for AI-Generated Image Detection with Vision-Language Large Models
- CVPR (T1)UDAPose: Unsupervised Domain Adaptation for Low-Light Human Pose Estimation
20259 papers
- NeurIPS (T1)3DOT: Texture Transfer for 3D Objects from a Single Reference Image
- NeurIPS (T1)GeoComplete: Geometry-Aware Diffusion for Reference-Driven Image Completion
- ICCV (T1)HOLa: Zero-Shot HOI Detection with Low-Rank Decomposed VLM Feature Adaptation
- CVPR (T1)Mamba as a Bridge: Where VFM Meets VLM for Domain-Generalized Semantic Segmentation
- AAAI (T1)NightHaze: Nighttime Image Dehazing via Self-Prior Learning
- AAAI Conference on Artificial Intelligence, AAAISemantic Segmentation on Raindrop Degraded Images using Two-Stage Dual Teacher-Student Learning
- BMVCSSNeRF: Sparse View Semi-supervised Neural Radiance Fields with Augmentation
- CVPR (T1)Synthetic-to-Real Self-supervised Robust Depth Estimation via Learning with Motion and Structure Priors
- Association for Computational Linguistics, ACLuMedSum: A Unified Framework for Clinical Abstractive Summarization
202414 papers
- CVPR (T1)CAT: Exploiting Inter-Class Dynamics for Domain Adaptive Object Detection
- AAAI (T1)DeS3: Adaptive Attention-driven Self and Soft Shadow Removal using ViT Similarity
- ECCV (T1)Domain-Adaptive 2D Human Pose Estimation via Dual Teachers in Extremely Low-Light Conditions
- ECCV (T1)Dual-Rain: Video Rain Removal using Assertive and Gentle Teachers
- NeurIPS (T1)End-to-End Video Semantic Segmentation in Adverse Weather using Fusion Blocks and Temporal-Spatial Teacher-Student Learning
- NeurIPS (T1)EZ-HOI: VLM Adaptation via Guided Prompt Learning for Zero-Shot HOI Detection
- AAAI (T1)Few-Shot Learning from Augmented Label-uncertain Queries in Bongard-HOI
- AAAI (T1)HEAP: Unsupervised Object Discovery and Localization with Contrastive Grouping
- CVPR (T1)NightCC: Nighttime Color Constancy via Adaptive Channel Masking
- AAAI (T1)NightRain: Nighttime Video Deraining via Self-Rain-Removal and Self-Correction
- GastroenterologyRapid Endoscopic Diagnosis of Benign Ulcerative Colorectal Diseases with an Artificial Intelligence Contextual Framework
- AAAI (T1)Restoring Speaking Lips from Occlusion for Audio-Visual Speech Recognition
202310 papers
- CVPR (T1)2PCNet: Two-Phase Consistency Training for Day-to-Night Unsupervised Domain Adaptive Object Detection
- ICCV (T1)Auxiliary Tasks Benefits 3D Skeleton-based Human Motion Prediction
- Pattern RecognitionBottom-Up 2D Pose Estimation via Dual Anatomical Centers for Small-Scale Persons
- ICCV (T1)Deep Homography Mixture for Single Image Rolling Shutter Correction
- CVPR (T1)DSFNet: Dual Space Fusion Network for Occlusion-Robust 3D Dense Face Alignment
- ACL (T1)Dynamic Transformers Provide a False Sense of Efficiency
- ACM MMEnhancing Visibility in Nighttime Haze Images Using Guided APSF and Gradient Adaptive Convolution
- CVPR (T1)EqMotion: Equivariant Multi-agent Motion Prediction with Invariant Interaction Reasoning
- AAAI (T1)Estimating Reflectance Layer from A Single Image: Integrating Reflectance Guidance and Shadow/Specular Aware Learning
- CVPR (T1)Seeing What You Said: Talking Face Generation Guided by a Lip Reading Expert
Where alumni went44 listed
Apply to NUSFall 2027
School of Computing PhD in Computer Science
- Round
- Main
- Opens
- —
- GRE
- See program page
- English
- See program page
- Fee
- See program page
Not yet confirmed for this intake.
Similar labsnearest by research summary
- CNSJTU
Guangtao Zhai
- Video understanding
- CNPeking
Yuxin Peng
- Recognition, detection & segmentation
- USUMD
Ruohan Gao
- Recognition, detection & segmentation
- HKCUHK
Hongsheng Li
- Vision-language & multimodal
- CNSJTU
Xiongkuo Min
- Video understanding
- CNPeking
Shiliang Zhang
- Recognition, detection & segmentation
Your notes
Save this lab to keep notes.
Data
Checked against the lab homepage.
- Collected
- Verified
- Papers
- The lab homepage
- Members
- Lab homepage
- Alumni
- Lab homepage
Tiers are venue classifications, not lab rankings.
NUScloses