Salman Khan

A new method improves landmark detection without human labeling.

A new method allows models to recognize both known and unknown objects.

MobiLlama offers efficient language processing for devices with limited resources.

Understanding model robustness is key for real-world applications in various fields.

New framework improves video searches by combining visuals and detailed language descriptions.

MAVOS introduces an efficient method for tracking objects in long video clips.

ELGC-Net improves accuracy in detecting changes using satellite images.

Learn to classify objects using images and 3D point clouds without labels.

Assessing the capabilities and challenges of advanced video understanding models.

Open-YOLO 3D enhances 3D instance segmentation with speed and accuracy.

Study examines the robustness of segmentation models against adversarial attacks in healthcare.

A new model enhances video comprehension by merging image and video encoders.

This article examines how Visual State Space Models handle visual challenges.

VANE-Bench enhances detection of anomalies in videos amidst growing AI content.

A new method improves video action recognition using contextual language.

CPT improves black-box model performance without direct access to internal parameters.

FANet enhances semantic segmentation, improving accuracy in complex images.

GroupMamba enhances image processing efficiency and accuracy in computer vision tasks.

Effective techniques to detect plastic waste in our oceans.

New methods expose vulnerabilities in medical models through backdoor attacks.

This study explores innovative ways to influence and interact with dreams through brain signals.

A study on improving weather predictions in the Middle East and North Africa.

New tools improve how we describe changes in satellite images over time.

ROAD-Waymo enhances understanding of road actions for autonomous vehicles.

VideoGLaMM enhances video understanding through detailed visual and textual connections.

GEOBench-VLM evaluates models for interpreting geospatial data and images.

A bilingual model transforming medical communication for patients and professionals.

RHFL+ tackles data noise and model differences in federated learning.

A new dataset revolutionizes analysis of medical images and their descriptions.

Simplifying environmental data through engaging conversations.