CamReasoner uses structured O-T-A reasoning and RL on 56k samples to lift camera movement classification from 73.8% to 78.4% and VQA from 60.9% to 74.5% on Qwen2.5-VL-7B.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2verdicts
UNVERDICTED 2representative citing papers
ProBA replaces rigid point tracks with a probabilistic pose graph and 3D Gaussian landmarks, optimizing via negative log-likelihood with the Bhattacharyya coefficient to expand the basin of attraction in prior-free SfM.
citing papers explorer
-
CamReasoner: Reinforcing Camera Movement Understanding via Structured Spatial Reasoning
CamReasoner uses structured O-T-A reasoning and RL on 56k samples to lift camera movement classification from 73.8% to 78.4% and VQA from 60.9% to 74.5% on Qwen2.5-VL-7B.
-
ProBA: Probabilistic Bundle Adjustment with the Bhattacharyya Coefficient
ProBA replaces rigid point tracks with a probabilistic pose graph and 3D Gaussian landmarks, optimizing via negative log-likelihood with the Bhattacharyya coefficient to expand the basin of attraction in prior-free SfM.