Picture for Dongbo Min

Dongbo Min

Senior Member, IEEE

Hybrid-TTA: Continual Test-time Adaptation via Dynamic Domain Shift Detection

Add code
Sep 13, 2024
Viaarxiv icon

iConFormer: Dynamic Parameter-Efficient Tuning with Input-Conditioned Adaptation

Add code
Sep 04, 2024
Viaarxiv icon

SG-MIM: Structured Knowledge Guided Efficient Pre-training for Dense Prediction

Add code
Sep 04, 2024
Viaarxiv icon

UniTT-Stereo: Unified Training of Transformer for Enhanced Stereo Matching

Add code
Sep 04, 2024
Figure 1 for UniTT-Stereo: Unified Training of Transformer for Enhanced Stereo Matching
Figure 2 for UniTT-Stereo: Unified Training of Transformer for Enhanced Stereo Matching
Figure 3 for UniTT-Stereo: Unified Training of Transformer for Enhanced Stereo Matching
Figure 4 for UniTT-Stereo: Unified Training of Transformer for Enhanced Stereo Matching
Viaarxiv icon

MaDis-Stereo: Enhanced Stereo Matching via Distilled Masked Image Modeling

Add code
Sep 04, 2024
Viaarxiv icon

CLDA: Collaborative Learning for Enhanced Unsupervised Domain Adaptation

Add code
Sep 04, 2024
Viaarxiv icon

Dynamic Guidance Adversarial Distillation with Enhanced Teacher Knowledge

Add code
Sep 03, 2024
Viaarxiv icon

Fine-grained Background Representation for Weakly Supervised Semantic Segmentation

Add code
Jun 22, 2024
Viaarxiv icon

Salience-Based Adaptive Masking: Revisiting Token Dynamics for Enhanced Pre-training

Add code
Apr 12, 2024
Viaarxiv icon

Emerging Property of Masked Token for Effective Pre-training

Add code
Apr 12, 2024
Viaarxiv icon