Search papers, labs, and topics across Lattice.
SGIT AI Lab, State Grid Corporation of China
1
0
3
Achieve strong spatio-temporal video grounding with only 10M trainable parameters by smartly adapting pre-trained 2D visual-language models.
