Open to Collab

7 135 276

Muhammad Umair

umair894

AI & ML interests

Multimodal Reidentification | Feature Upscaling | Object Tracking |PhD UESTC

Recent Activity

upvoted an article 5 days ago

SmolVLA: Efficient Vision-Language-Action Model trained on Lerobot Community Data

upvoted a paper 5 days ago

UPLiFT: Efficient Pixel-Dense Feature Upsampling with Local Attenders

liked a Space 6 days ago

Overworld/waypoint-1-small

View all activity

Organizations

liked a Space 6 days ago

Waypoint 1 Small

🎮

Explore and navigate through AI-generated worlds in real-time

liked 2 Spaces 11 days ago

SAMTok

🎨

SAMTok provides a unified mask-token interface for MLLMs.

Dream-wan2-2-faster-Pro

🎥

1.22k

generate a video from an image with a text prompt

liked a Space 13 days ago

2025 China AI Timeline

📈

liked a Space 27 days ago

Qwen Edit Any Pose

🕺

194

Edit any pose with Qwen Edit 2511 Any Pose LoRA

liked a Space 29 days ago

HY-Motion-1.0

💃

221

Text-to-3D and Image-to-3D Generation

liked 2 Spaces about 1 month ago

Mapanything Gradio

🐠

115

Convert images to 3D models and depth maps

WindowSeat Reflection Removal

🪟

Remove reflections from images

liked a model about 2 months ago

google/functiongemma-270m-it

Text Generation • 0.3B • Updated 20 days ago • 90.1k • 867

liked 4 Spaces about 2 months ago

CUGA Agent

🤖

Configurable Generalist Agent, leader in AppWorld Benchmark

OutofLipSync

💋

Translate and Lipsync any video clips to English

AnyTalker

🎬

Let your character interact naturally

Anim Lab AI

🔥

129

Turn boring math into engaging animation videos

liked a model about 2 months ago

microsoft/VibeVoice-Realtime-0.5B

Text-to-Speech • 1B • Updated Dec 12, 2025 • 323k • 1.1k

liked a model 2 months ago

IDEA-Research/RexSeek-3B

Image-Text-to-Text • 4B • Updated Mar 14, 2025 • 26 • 10

liked 4 Spaces 2 months ago

SAM3D Body with Rerun

🧍

SAM3 Video Segmentation

🐠

105

Track and label objects in videos using text prompts or clicks

CUA - Computer Use Agent 2.0

🤖

125

Generate captions for images

EdgeTAM

🚀

On-Device Track Anything Model

liked a Space 3 months ago

Miragic Virtual Try On

👕

504

Try on complete outfits with our virtual try-on technology