Publications

Image-Guided Semantic Pseudo-LiDAR Point Generation for 3D Object Detection

Style-Friendly SNR Sampler for Style-Driven Generation

Towards Failure-Resilient Lifelong Learning Agents through Scene Graph-Guided Proactive Replanning

Know "No" Better: A Data-Driven Approach for Enhancing Negation Awareness in CLIP

Watermarking for Factuality: Guiding Vision-Language Models Toward Truth via Tri-layer Contrastive Decoding

SemanticControl: A Training-free Approach for Handling Loosely Aligned Visual Conditions in ControlNet

DiffExp: Efficient Exploration in Reward Fine-tuning for Text-to-Image Diffusion Models

3D Occupancy Prediction with Low-Resolution Queries via Prototype-aware View Transformation

GUIDE-CoT: Goal-driven and User-Informed Dynamic Estimation for Pedestrian Trajectory using Chain-of-Thought

ViTA-PAR: Visual and Textual Attribute Alignment with Attribute Prompting for Pedestrian Attribute Recognition

Hard-Negative Prototype-Based Regularization for Few-Shot Class-Incremental Learning

Encoder-only Next Token Prediction

Fast and Precise Multimodal Spatiotemporal Calibration via Periodic-Activated 2D Gaussian Splatting

Parameter-Efficient 12-Lead ECG Reconstruction from a Single Lead

Fpanet: Frequency-based video demoireing using frame-level post alignment

Unified Domain Generalization and Adaptation for Multi-View 3D Object Detection

Toward Interactive Regional Understanding in Vision-Large Language Models

VisionTrap: Vision-Augmented Trajectory Prediction Guided by Textual Descriptions

Higher-order Relational Reasoning for Pedestrian Trajectory Prediction

Audio-guided implicit neural representation for local image stylization

Robust sound-guided image manipulation

CMDA: Cross-Modal and Domain Adversarial Adaptation for LiDAR-Based 3D Object Detection

InstructBooth: Instruction-following Personalized Text-to-Image Generation

Improving Visual Prompt Tuning for Self-supervised Vision Transformers

Weakly Supervised Referring Image Segmentation with Intra-Chunk and Inter-Chunk Consistency

CloudNet: A LiDAR-based face anti-spoofing model that is robust against light variation

An Embedding-Dynamic Approach to Self-Supervised Learning

Anti-Adversarially Manipulated Attributions for Weakly Supervised Semantic Segmentation and Object Localization

Weakly Supervised Semantic Segmentation using Out-of-Distribution Data

Perception Prioritized Training of Diffusion Models

Bridging the Gap between Classification and Localization for Weakly Supervised Object Localization

Ora3d: Overlap region aware multi-view 3d object detection

Stopnet: Scalable trajectory and occupancy prediction for urban autonomous driving

Occupancy flow fields for motion forecasting in autonomous driving

Reducing Information Bottleneck for Weakly Supervised Semantic Segmentation

Toward Spatially Unbiased Generative Models

Anti-Adversarially Manipulated Attributions for Weakly and Semi-Supervised Semantic Segmentation

BBAM: Bounding Box Attribution Map for Weakly Supervised Semantic and Instance Segmentation

Selfreg: Self-supervised contrastive regularization for domain generalization

Advisable Learning for Self-Driving Vehicles by Internalizing Observation-to-Action Rules

Frame-to-Frame Aggregation of Active Regions in Web Videos for Weakly Supervised Semantic Segmentation

FickleNet: Weakly and Semi-supervised Semantic Image Segmentation using Stochastic Inference

Grounding Human-to-Vehicle Advice for Self-driving Vehicles

Msnet: Mutual suppression network for disentangled video representations

Textual Explanations for Self-Driving Vehicles

Deep traffic light detection for self-driving cars from a large-scale dataset

Interpretable Learning for Self-driving Cars by Visualizing Causal Attention