Diana Marculescu

8

Papers

0

Total Citations

Papers (8)

Movie Weaver: Tuning-Free Multi-Concept Video Personalization with Anchored Prompts

FlowVid: Taming Imperfect Optical Flows for Consistent Video-to-Video Synthesis

T-VSL: Text-Guided Visual Sound Source Localization in Mixtures

Regularizing Activation Distribution for Training Binarized Deep Networks

Towards Efficient Model Compression via Learned Global Ranking

Open-Vocabulary Semantic Segmentation With Mask-Adapted CLIP

Jumping through Local Minima: Quantization in the Loss Landscape of Vision Transformers

Efficient Low-rank Backpropagation for Vision Transformer Adaptation