DiNO Mix

DINO (DETR with Improved DeNoising Anchor Boxes) and its various extensions represent a family of self-supervised learning methods primarily focused on improving visual representation learning for various computer vision tasks. Current research emphasizes adapting DINO's powerful feature extraction capabilities to diverse applications, including medical image analysis, open-vocabulary object detection, and robot manipulation, often through techniques like feature mixing and adapter learning. This work is significant because it demonstrates the potential of self-supervised learning to address data scarcity issues in specialized domains and improve the efficiency and generalizability of computer vision models across a wide range of applications.

Papers

August 8, 2023

Temporal DINO: A Self-supervised Video Strategy to Enhance Action Prediction
Izzeddin Teeti, Rongali Sai Bhargav, Vivek Singh, Andrew Bradley, Biplab Banerjee, Fabio Cuzzolin
LSTM Network Action Prediction DiNO Mix Video Task Self Supervised Video

August 1, 2023

DINO-CXR: A self supervised method based on vision transformer for chest X-ray classification
Mohammadreza Shakouri, Fatemeh Iranmanesh, Mahdi Eftekhari
Vision Transformer Practical Method Medical Image Analysis Self Supervised Pretraining Self Information Chest X Ray Datasets DiNO Mix Chest X Ray Classification

May 17, 2023

DinoSR: Self-Distillation and Online Clustering for Self-supervised Speech Representation Learning
Alexander H. Liu, Heng-Jui Chang, Michael Auli, Wei-Ning Hsu, James R. Glass
Self Distillation Contextual Embeddings Self Supervised Speech Representation Learning DiNO Mix Online Clustering Representation Learner

April 6, 2023

From Saliency to DINO: Saliency-guided Vision Transformer for Few-shot Keypoint Detection
Changsheng Lu, Hao Zhu, Piotr Koniusz
Saliency Map Human Saliency Keypoint Detector DiNO Mix Saliency Guided Supervised Keypoint

March 9, 2023

Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection
Shilong Liu, Zhaoyang Zeng, Tianhe Ren, Feng Li, Hao Zhang, Jie Yang, Chunyuan Li, Jianwei Yang, Hang Su, Jun Zhu, Lei Zhang
Pre Training Open Set Open Set Object DiNO Mix Open Domain Generalization

August 15, 2022

C3-DINO: Joint Contrastive and Non-contrastive Self-Supervised Learning for Speaker Verification
Chunlei Zhang, Dong Yu
Contrastive Learning Self Supervised Speaker Verification Speaker Embeddings Contrastive Self Supervised Learning Joint Framework DiNO Mix Non Contrastive Self Supervised Learning Discriminative Speaker

June 14, 2022

Exploring Adversarial Attacks and Defenses in Vision Transformers trained with DINO
Javier Rando, Nasib Naimi, Thomas Baumann, Max Mathys
Adversarial Attack Vision Transformer Adversarial Training Adversarial Robustness Self Supervised Vision Transformer DiNO Mix Ensemble Adversarial

June 6, 2022

Mask DINO: Towards A Unified Transformer-based Framework for Object Detection and Segmentation
Feng Li, Hao Zhang, Huaizhe xu, Shilong Liu, Lei Zhang, Lionel M. Ni, Heung-Yeung Shum
Semantic Segmentation Object Detection Segmentation Based Approach Unified Alignment Transformer Based Framework Mask Prediction DiNO Mix Segmentation Framework

March 7, 2022

DINO: DETR with Improved DeNoising Anchor Boxes for End-to-End Object Detection
Hao Zhang, Feng Li, Shilong Liu, Lei Zhang, Hang Su, Jun Zhu, Lionel M. Ni, Heung-Yeung Shum
End to End Multi Scale Feature Multi Object Object Box Group DETR V2 DiNO Mix

DiNO Mix

Papers

Temporal DINO: A Self-supervised Video Strategy to Enhance Action Prediction

DINO-CXR: A self supervised method based on vision transformer for chest X-ray classification

DinoSR: Self-Distillation and Online Clustering for Self-supervised Speech Representation Learning

From Saliency to DINO: Saliency-guided Vision Transformer for Few-shot Keypoint Detection

Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection

C3-DINO: Joint Contrastive and Non-contrastive Self-Supervised Learning for Speaker Verification

Exploring Adversarial Attacks and Defenses in Vision Transformers trained with DINO

Mask DINO: Towards A Unified Transformer-based Framework for Object Detection and Segmentation

DINO: DETR with Improved DeNoising Anchor Boxes for End-to-End Object Detection