Article

KTVIC: A Vietnamese Image Captioning Dataset on the Life Domain

Introduces KTVIC, a life-domain Vietnamese image captioning dataset for benchmarking multimodal models.

Anh-Cuong Pham
•
Read more

Visual Abductive Reasoning Meets Driving Hazard Prediction: Problem Formulation and Dataset

Defines a visual abductive reasoning benchmark for hazard prediction with a newly collected dataset.

Korawat Charoenpitaks
•
Read more

Leveraging Video Coding Knowledge for Deep Video Enhancement

Transfers video coding priors into deep enhancement networks for improved restoration quality.

Thong Bach
•
Read more