InstanceControl uses VLMs to auto-generate instance masks from text and visual conditions, with adaptive refinement, to enable controllable multi-object image generation without manual labeling.
arXiv preprint arXiv:2502.10451 (2025)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
InstanceControl: Controllable Complex Image Generation without Instance Labeling
InstanceControl uses VLMs to auto-generate instance masks from text and visual conditions, with adaptive refinement, to enable controllable multi-object image generation without manual labeling.