-
Direct-a-Video: Customized Video Generation with User-Directed Camera Movement and Object Motion
Paper • 2402.03162 • Published • 17 -
ShortGPT: Layers in Large Language Models are More Redundant Than You Expect
Paper • 2403.03853 • Published • 61 -
OpenVid-1M: A Large-Scale High-Quality Dataset for Text-to-video Generation
Paper • 2407.02371 • Published • 51
Pengxiang Li
pengxiang
AI & ML interests
Video generation, Image editing, AD
Recent Activity
upvoted
a
paper
about 4 hours ago
rStar-Math: Small LLMs Can Master Math Reasoning with Self-Evolved Deep
Thinking
authored
a paper
2 days ago
TrackDiffusion: Tracklet-Conditioned Video Generation via Diffusion
Models
authored
a paper
2 days ago
Automated Evaluation of Large Vision-Language Models on Self-driving
Corner Cases
Organizations
None yet
Collections
1
models
6
pengxiang/TrackDiffusion_SVD_Stage2
Text-to-Video
•
Updated
pengxiang/TrackDiffusion_SVD_Stage1
Text-to-Video
•
Updated
pengxiang/TrackDiffusion_Pretrain
Updated
•
3
•
1
pengxiang/GLIGEN_1_4
Updated
•
5
pengxiang/TrackDiffusion_ModelScope
Text-to-Video
•
Updated
pengxiang/trackdiffusion_ytvis
Text-to-Video
•
Updated
•
2