Stars
Seed1.5-VL, a vision-language foundation model designed to advance general-purpose multimodal understanding and reasoning, achieving state-of-the-art performance on 38 out of 60 public benchmarks.
MambaOut: Do We Really Need Mamba for Vision? (CVPR 2025)
Scaling Up Your Kernels to 31x31: Revisiting Large Kernel Design in CNNs (CVPR 2022)
[CVPR'24] UniRepLKNet: A Universal Perception Large-Kernel ConvNet for Audio, Video, Point Cloud, Time-Series and Image Recognition
ACNet: Strengthening the Kernel Skeletons for Powerful CNN via Asymmetric Convolution Blocks
InteractiveVideo: User-Centric Controllable Video Generation with Synergistic Multimodal Instructions
[CVPR'24] Multimodal Pathway: Improve Transformers with Irrelevant Data from Other Modalities
Unsupervised deep learning framework with online(MLP: prediction-based, 1 D Conv and VAE: reconstruction-based, Wavenet: prediction-based) settings for anaomaly detection in time series data
[NeurIPS'22] This is an official implementation for "Scaling & Shifting Your Features: A New Baseline for Efficient Model Tuning".
Diverse Branch Block: Building a Convolution as an Inception-like Unit
Keras (TensorFlow v2) reimplementation of Re-parameterized Large Kernel Network (RepLKNet)
RepMLPNet: Hierarchical Vision MLP with Re-parameterized Locality (CVPR 2022)
Is the attention layer even necessary? (https://arxiv.org/abs/2105.02723)
TensorRT implementation of "RepVGG: Making VGG-style ConvNets Great Again"
ResRep: Lossless CNN Pruning via Decoupling Remembering and Forgetting (ICCV 2021)