HMAT restores murals more coherently and faithfully than prior methods by integrating mask-aware dynamic filtering, a transformer bottleneck, style fusion, and gated skip connections.
In: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2years
2026 2representative citing papers
citing papers explorer
-
High-Fidelity Mural Restoration via a Unified Hybrid Mask-Aware Transformer
HMAT restores murals more coherently and faithfully than prior methods by integrating mask-aware dynamic filtering, a transformer bottleneck, style fusion, and gated skip connections.
- Anatomy-Slot: Unsupervised Anatomical Factorization for Homologous Bilateral Reasoning in Retinal Diagnosis