Armstrong Aboah

Stop treating traffic forecasting as just image reconstruction: explicitly modeling road structure yields significantly more accurate and transferable traffic predictions across cities.

Achieve near-perfect traffic congestion classification by fusing motion-guided visual attention with data-adaptive temporal decomposition, outperforming existing vision-based and signal-based methods.

Domain-specific instruction tuning can transform general vision-language models into expert systems capable of comprehensive and standardized infrastructure assessment.