A training-free framework that organizes CLIP-based dense prediction around segment-level region abstractions, reporting state-of-the-art mIoU on eight open-vocabulary segmentation benchmarks.
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
SCI-CLIP: Segment-Centric Inference with Reference Memory for Training-Free Open-Vocabulary Segmentation
A training-free framework that organizes CLIP-based dense prediction around segment-level region abstractions, reporting state-of-the-art mIoU on eight open-vocabulary segmentation benchmarks.