Shariq Farooq
@shariq_farooq
Followers
341
Following
1K
Media
7
Statuses
228
PhD @ KAUST👨💻 | Worked with @intel , @Adobe , @Stanford @ucl | Author of ZoeDepth, AdaBins, LooseControl | Exploring the 3D world, one image at a time 🔍 |
Srinagar, Kashmir
Joined May 2012
Our work, ZoeDepth, has been **rejected** from ICCV. But I am absolutely thrilled at how well it was received by the community!. Here are some of my favorite use cases 👇.
17
41
271
RT @AbdalRameen: What if you could compose videos— merging multiple clips, even capturing complex athletic moves where video models struggl….
0
25
0
RT @SchmidhuberAI: I am hiring 3 postdocs at #KAUST to develop an Artificial Scientist for discovering novel chemical materials for carbon….
0
68
0
RT @OpenAI: Introducing Sora, our text-to-video model. Sora can create videos of up to 60 seconds featuring highly detailed scenes, comple….
0
31K
0
RT @duborges: This video was generated by Sora. That's the new model by OpenAI. The most advanced text-to-video tool created so far. I'll….
0
2K
0
RT @PonchMichael: Cinematic Title Background: . 1) Prompt 3D texture in #midjourney .2) Clean BG in #photoshop .3) 2D animate in @AdobeVide….
0
61
0
RT @isskoro: This paper has received significantly less attention than it deserves, so let me shed a bit more light on it and describe why….
0
228
0
😍😍.
Need more control over your generations? Create the foreground and background separately and use ZoeDepth to create the 3D background. A quick tutorial here. I’ll be using this for my next project together with #makebigfilms
1
1
2
RT @Abdo_Eldesokey: Check out my most recent work with @peter_wonka : "Text2AC-Zero" for generating videos of Animated Characters using pre….
0
4
0
PatchFusion lifts ZoeDepth to very high resolution and natively supports 4K+ input! . Can’t wait to see the community trying it out for various applications where ZoeDepth couldn’t provide the details!.
PatchFusion: An End-to-End Tile-Based Framework for High-Resolution Monocular Metric Depth Estimation @Gradio demo is out on @huggingface . demo: run with docker: duplicate space with private gpu and no queue:
0
3
27
RT @Gradio: Two excellent Depth estimation models have recently come out-.1⃣PatchDiffusion (@zhenyu_li9955,@shariq_farooq et al.).2⃣Marigol….
0
13
0
RT @80Level: Researchers presented LooseControl, a model that can change AI-generated images by using 3D boxes. You can set up the boxes, a….
0
80
0
RT @alexcarliera: PatchFusion was just released. Compared to ZoeDepth, it predicts depth maps with much finer details, just look at the com….
0
128
0
RT @alexcarliera: LooseControl was just released and it's so good! 🔥. It enables depth-map conditioned image generation, but unlike Control….
0
112
0
Of course, there is a long way to go. The source code will be released soon. I hope one day this matures enough to be the go-to choice for 3D design within Gradio.
0
1
5
PatchFusion, our new 4K depth estimation model, significantly improves the details and quality of depth maps produced by ZoeDepth 👇.
Check our new work PatchFusion! It estimates high-resolution metric depth with intricate details for high-resolution input image. Zhenyu Li, Shariq Farooq Bhat (@shariq_farooq), Peter Wonka (@peter_wonka). Webpage:
0
8
31
@ak LooseControl can prove to be a new way to design complex scenes and perform semantic editing e.g. Model understands how lighting changes with the edits: (2/2)
1
10
71
Thanks @AK for featuring #LooseControl.Project page: TLDR: We extend ControlNet to new types of conditions that are easy to design by user. This opens up vast number of possibilities. e.g a user can manipulate objects in an image in a 3D-aware manner (1/2).
LooseControl: Lifting ControlNet for Generalized Depth Conditioning. paper page: present LooseControl to allow generalized depth conditioning for diffusion-based image generation. ControlNet, the SOTA for depth-conditioned image generation, produces
3
15
66