מושגי ליבה
提案されたDisentangled Object-Centric TRansformer(DOCTR)は、複数のオブジェクトと複数のサブタスクを統一的に学習することを可能にします。
סטטיסטיקה
Qualitative experimental results demonstrate that our method achieves state-of-the-art performance on the challenging ScanNet dataset.
ציטוטים
"Our method enables direct sparse predictions, yielding fewer false positives compared to existing methods."
"Extensive experiments demonstrate our superior performance than previous SOTA methods."