RoMa sets new state-of-the-art dense feature matching performance by fusing DINOv2 features with local ConvNet features, using anchor-probability transformer decoding, and regression-by-classification loss, with a 36% gain on WxBS.
Dgc-net: Dense ge- ometric correspondence network
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2verdicts
UNVERDICTED 2representative citing papers
Normalized Matching Transformer enforces unit-norm embeddings at every Transformer layer and trains with InfoNCE plus hyperspherical uniformity loss, reaching new state-of-the-art accuracy on PascalVOC and SPair-71k while converging faster than prior matching networks.
citing papers explorer
-
RoMa: Robust Dense Feature Matching
RoMa sets new state-of-the-art dense feature matching performance by fusing DINOv2 features with local ConvNet features, using anchor-probability transformer decoding, and regression-by-classification loss, with a 36% gain on WxBS.
-
Normalized Matching Transformer
Normalized Matching Transformer enforces unit-norm embeddings at every Transformer layer and trains with InfoNCE plus hyperspherical uniformity loss, reaching new state-of-the-art accuracy on PascalVOC and SPair-71k while converging faster than prior matching networks.