Guanglu Song

20

Papers

88

Total Citations

Papers (20)

FouriScale: A Frequency Perspective on Training-Free High-Resolution Image Synthesis

Rethinking the Spatial Inconsistency in Classifier-Free Diffusion Guidance

See Further When Clear: Curriculum Consistency Model

Revisiting the Sibling Head in Object Detector

Switchable K-Class Hyperplanes for Noise-Robust Representation Learning

DETRs with Collaborative Hybrid Assignments Training

Decoupled DETR: Spatially Disentangling Localization and Classification for Improved End-to-End Object Detection

Masked Autoencoders Are Stronger Knowledge Distillers

Temporal Enhanced Training of Multi-view 3D Object Detector via Historical Object Prediction

Discriminability Distillation in Group Representation Learning

Unifying Visual Perception by Dispersible Points Learning

Self-Slimmed Vision Transformer

Rethinking Robust Representation Learning under Fine-Grained Noisy Faces

Towards Robust Face Recognition with Comprehensive Search

"UniNet: Unified Architecture Search with Convolution, Transformer, and MLP"

UniKD: Universal Knowledge Distillation for Mimicking Homogeneous or Heterogeneous Object Detectors

LMDrive: Closed-Loop End-to-End Driving with Large Language Models

Beyond Trade-Off: Accelerate FCN-Based Face Detector With Higher Accuracy

Large-batch Optimization for Dense Visual Predictions: Training Faster R-CNN in 4.2 Minutes

RAPHAEL: Text-to-Image Generation via Large Mixture of Diffusion Paths