Picture for Dongfang Liu

Dongfang Liu

EAVE: Efficient Product Attribute Value Extraction via Lightweight Sparse-layer Interaction

Add code
Jun 10, 2024
Viaarxiv icon

Self-supervised Adversarial Training of Monocular Depth Estimation against Physical-World Attacks

Add code
Jun 09, 2024
Viaarxiv icon

ProMotion: Prototypes As Motion Learners

Jun 07, 2024
Viaarxiv icon

Prototypical Transformer as Unified Motion Learners

Add code
Jun 03, 2024
Viaarxiv icon

SSGA-Net: Stepwise Spatial Global-local Aggregation Networks for for Autonomous Driving

May 29, 2024
Viaarxiv icon

BadPart: Unified Black-box Adversarial Patch Attacks against Pixel-wise Regression Tasks

Apr 01, 2024
Viaarxiv icon

Text Is MASS: Modeling as Stochastic Embedding for Text-Video Retrieval

Add code
Mar 26, 2024
Figure 1 for Text Is MASS: Modeling as Stochastic Embedding for Text-Video Retrieval
Figure 2 for Text Is MASS: Modeling as Stochastic Embedding for Text-Video Retrieval
Figure 3 for Text Is MASS: Modeling as Stochastic Embedding for Text-Video Retrieval
Figure 4 for Text Is MASS: Modeling as Stochastic Embedding for Text-Video Retrieval
Viaarxiv icon

Image Translation as Diffusion Visual Programmers

Add code
Jan 30, 2024
Viaarxiv icon

Facing the Elephant in the Room: Visual Prompt Tuning or Full Finetuning?

Add code
Jan 23, 2024
Viaarxiv icon

Efficient Multimodal Semantic Segmentation via Dual-Prompt Learning

Add code
Dec 04, 2023
Viaarxiv icon