Xin Tan

22

Papers

258

Total Citations

Papers (22)

PromptAD: Learning Prompts with only Normal Samples for Few-Shot Anomaly Detection

COTR: Compact Occupancy TRansformer for Vision-based 3D Occupancy Prediction

DrivingForward: Feed-forward 3D Gaussian Splatting for Driving Scene Reconstruction from Flexible Surround-view Input

FastLGS: Speeding Up Language Embedded Gaussians with Feature Grid Mapping

Continuous Piecewise-Affine Based Motion Model for Image Animation

Beyond the Label Itself: Latent Labels Enhance Semi-supervised Point Cloud Panoptic Segmentation

MOS: Modeling Object-Scene Associations in Generalized Category Discovery

Rethinking Efficient Lane Detection via Curve Modeling

Learning To Detect Mirrors From Videos via Dual Correspondences

Boosting Semi-Supervised Learning by Exploiting All Unlabeled Data

Multi-Centroid Task Descriptor for Dynamic Class Incremental Inference

Rethinking Gradient Projection Continual Learning: Stability / Plasticity Feature Space Decoupling

Re-ID Driven Localization Refinement for Person Search

Instance and Category Supervision are Alternate Learners for Continual Learning

One-for-More: Continual Diffusion Model for Anomaly Detection

Optimization over Disentangled Encoding: Unsupervised Cross-Domain Point Cloud Completion via Occlusion Factor Manipulation

Domain-Hallucinated Updating for Multi-Domain Face Anti-spoofing

Learning Task-Aware Language-Image Representation for Class-Incremental Object Detection

A Two-Stage Information Extraction Network for Incomplete Multi-View Multi-Label Classification

Building a Strong Pre-Training Baseline for Universal 3D Large-Scale Perception

Multi-modal In-Context Learning Makes an Ego-evolving Scene Text Recognizer

Omni-Supervised Point Cloud Segmentation via Gradual Receptive Field Component Reasoning