Search papers, labs, and topics across Lattice.
The University of Hong Kong
2
0
3
Achieving a 75% reduction in inference latency while maintaining high reconstruction quality could revolutionize how we deploy Implicit Neural Representations in real-time applications.
By pruning and quantizing the KV cache, XStreamVGGT achieves a remarkable 4.42x memory reduction and 5.48x speedup in streaming 3D reconstruction without sacrificing performance.
