Abstract: We study few-shot fine-grained image classification, a task that faces two key challenges: (1) the scarcity of labeled samples amplifies the model’s sensitivity to viewpoint variations, ...

Hybrid web crawler that intelligently renders JavaScript-heavy pages only when needed — balances content coverage and infrastructure cost using heuristic scoring + conditional Playwright rendering.

We introduce MVInverse, aiming to address the limitations of existing methods—such as inconsistent results or high computational costs—when reconstructing scene geometry and materials from multiple ...

Abstract: 3D classification is complex and challenging because of high-dimensional data, the intricate nature of their spatial relationships, and viewpoint variations. We fill the gap in view-based 3D ...