SOTAVerified

Visual Prompting

Visual Prompting is the task of streamlining computer vision processes by harnessing the power of prompts, inspired by the breakthroughs of text prompting in NLP. This innovative approach involves using a few visual prompts to swiftly convert an unlabeled dataset into a deployed model, significantly reducing development time for both individual projects and enterprise solutions.

Papers

Showing 121127 of 127 papers

TitleStatusHype
Unleashing the Power of Visual Prompting At the Pixel LevelCode0
Towards Ambiguity-Free Spatial Foundation Model: Rethinking and Decoupling Depth AmbiguityCode0
Stepwise Decomposition and Dual-stream Focus: A Novel Approach for Training-free Camouflaged Object SegmentationCode0
Uncovering the Hidden Cost of Model CompressionCode0
Benchmarking Human and Automated Prompting in the Segment Anything ModelCode0
ViP-LLaVA: Making Large Multimodal Models Understand Arbitrary Visual PromptsCode0
Leveraging Large Language Models for Scalable Vector Graphics-Driven Image UnderstandingCode0
Show:102550
← PrevPage 13 of 13Next →

No leaderboard results yet.