Warning: mkdir(): No space left on device in /var/www/hottg/post.php on line 59

Warning: file_put_contents(aCache/aDaily/2025-07-22/post/datastorieslanguages/--): Failed to open stream: No such file or directory in /var/www/hottg/post.php on line 72
​​Contrastive Feature Masking Open-Vocabulary Vision Transformer @Data, Stories and Languages
TG Telegram Group & Channel
Data, Stories and Languages | United States America (US)
Create: Update:

​​Contrastive Feature Masking Open-Vocabulary Vision Transformer

Contrastive Feature Masking Vision Transformer (CFM-ViT): a new approach for image-text pretraining that is optimized for open-vocabulary object detection. Unlike traditional masked autoencoders, which typically operate in the pixel space, CFM-ViT uses a joint image-text embedding space for reconstruction. This approach enhances the model's ability to learn region-level semantics. Additionally, the model features a Positional Embedding Dropout to better handle scale variations that occur when transitioning from image-text pretraining to detection finetuning. PED also enables the model to use a "frozen" ViT backbone as a region classifier without loss of performance.

In terms of results, CFM-ViT sets a new benchmark in open-vocabulary object detection with a 33.9 APr score on the LVIS dataset, outperforming the closest competitor by 7.6 points. The model also demonstrates strong capabilities in zero-shot detection transfer. Beyond object detection, it excels in image-text retrieval, outperforming the state of the art on 8 out of 12 key metrics. These features and results position CFM-ViT as a significant advancement in the field of computer vision and machine learning.

Paper link: https://arxiv.org/abs/2309.00775

My overview of the paper:
https://andlukyane.com/blog/paper-review-cfmvit
https://artgor.medium.com/paper-review-contrastive-feature-masking-open-vocabulary-vision-transformer-4639d1bf7043

#paperreview

​​Contrastive Feature Masking Open-Vocabulary Vision Transformer

Contrastive Feature Masking Vision Transformer (CFM-ViT): a new approach for image-text pretraining that is optimized for open-vocabulary object detection. Unlike traditional masked autoencoders, which typically operate in the pixel space, CFM-ViT uses a joint image-text embedding space for reconstruction. This approach enhances the model's ability to learn region-level semantics. Additionally, the model features a Positional Embedding Dropout to better handle scale variations that occur when transitioning from image-text pretraining to detection finetuning. PED also enables the model to use a "frozen" ViT backbone as a region classifier without loss of performance.

In terms of results, CFM-ViT sets a new benchmark in open-vocabulary object detection with a 33.9 APr score on the LVIS dataset, outperforming the closest competitor by 7.6 points. The model also demonstrates strong capabilities in zero-shot detection transfer. Beyond object detection, it excels in image-text retrieval, outperforming the state of the art on 8 out of 12 key metrics. These features and results position CFM-ViT as a significant advancement in the field of computer vision and machine learning.

Paper link: https://arxiv.org/abs/2309.00775

My overview of the paper:
https://andlukyane.com/blog/paper-review-cfmvit
https://artgor.medium.com/paper-review-contrastive-feature-masking-open-vocabulary-vision-transformer-4639d1bf7043

#paperreview
🔥1


>>Click here to continue<<

Data, Stories and Languages






Share with your best friend
VIEW MORE

United States America Popular Telegram Group (US)


Warning: Undefined array key 3 in /var/www/hottg/function.php on line 115

Fatal error: Uncaught mysqli_sql_exception: Too many connections in /var/www/db.php:16 Stack trace: #0 /var/www/db.php(16): mysqli_connect() #1 /var/www/hottg/function.php(212): db() #2 /var/www/hottg/function.php(115): select() #3 /var/www/hottg/post.php(351): daCache() #4 /var/www/hottg/route.php(63): include_once('...') #5 {main} thrown in /var/www/db.php on line 16