When do we not need larger vision models?

[ECCV 2024] Match-Stereo-Videos: Bidirectional Alignment for Consistent Dynamic Stereo Matching.

Cambrian-1 is a family of multimodal LLMs with a vision-centric design.

Efficient computing methods developed by Huawei Noah's Ark Lab

An open-source implementation of LLaVA-NeXT.

[CVPR 2024] VkD : Improving Knowledge Distillation using Orthogonal Projections

[CVPR 2024 - Oral] Matching 2D Images in 3D: Metric Relative Pose from Metric Correspondences

LLM training in simple, raw C/CUDA

A coding-free framework built on PyTorch for reproducible deep learning studies. 🏆25 knowledge distillation methods presented at CVPR, ICLR, ECCV, NeurIPS, ICCV, etc are implemented so far. 🎁 Train…

[ICML 2024] Vision Mamba: Efficient Visual Representation Learning with Bidirectional State Space Model

[ICCV 2023] Binary Adapters, [AAAI 2023] FacT, [Tech report] Convpass

[AAAI 2024] Understanding the Role of the Projector in Knowledge Distillation

Mamba SSM architecture

Implementation of the paper "Learning to Prompt CLIP for Monocular Depth Estimation: Exploring the Limits of Human Language", ICCV Workshop on Open Vocabulary Scene Understanding (OpenSUN3D) 2023

mPLUG-Owl & mPLUG-Owl2: Modularized Multimodal Large Language Model

🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.

Fine-tuning Vision Transformers on various classification datasets

The RedPajama-Data repository contains code for preparing large datasets for training large language models.

A playbook for systematically maximizing the performance of deep learning models.

Official DeiT repository

Official codebase for I-JEPA, the Image-based Joint-Embedding Predictive Architecture. First outlined in the CVPR paper, "Self-supervised learning from images with a joint-embedding predictive arch…

Official implementation of "Monocular Depth Estimation Using Cues Inspired by Biological Vision Systems", D. Auty and K. Mikolajczyk, International Conference on Pattern Recognition (ICPR) 2022

The repository provides code for running inference with the SegmentAnything Model (SAM), links for downloading the trained model checkpoints, and example notebooks that show how to use the model.

[BMVC 2022] Information Theoretic Representation Distillation