Tag
Introduces MultiDepth-3k, a benchmark to evaluate depth-layer preferences in monocular depth foundation models, and shows Laplacian Visual Prompting can alter reported depth layers, suggesting complementary geometric hypotheses exist across models.
Bardienus Duisterhof introduces Modality Forcing, a recipe for post-training text-to-image (T2I) models that achieves state-of-the-art results on 4 out of 5 monocular depth estimation benchmarks.