Weihan Xu PRO

WitneyWW

https://wx83.github.io/

wx83

AI & ML interests

None yet

Recent Activity

updated a Space 2 days ago

WitneyWW/pyramid-review

published a Space 2 days ago

WitneyWW/weekly-1008

updated a Space 2 days ago

WitneyWW/weekly-1008

View all activity

Organizations

None yet

WitneyWW 's Spaces 118

JavisDiT Edit — film adaln_sum step 12036 (test)

🎛

Browse video edit results with side‑by‑side comparisons

Audio VAE Roundtrip Highlighted

🔁

Compare original and VAE‑reconstructed audio clips

Audio Highlight New Vs Old

🔊

Generate enhanced audio highlights from video clips

VT world model - objective, force, flow, history (full test set x 3 seeds)

🔬

Explore performance of video‑tactile diffusion models

Where should the anchor frame go at inference?

⚓

Explore video generation anchor frame experiment results

Fixing the anchor - pinned vs free

📌

Explore video anchor comparison results with interactive visuals

JavisDiT Edit — caption finetune step 13000

💬

Explore video edit comparisons with captions and metrics

Conditioning arms for AV object editing

🎛

Explore AI methods for adding or removing objects in videos

JavisDiT Edit — audiofix scratch 0811 step 133500 (test)

🔊

Explore JavisDiT video/audio edit performance

JavisDiT Edit — step 133500 anchor x caption arms

🎯

Explore video edit results with anchor frames and captions

VT World Model Matrix20

🧪

View and compare visuo‑tactile model rollouts and metrics

World Frame Offset 0519

📐

Apply world-frame offset to dataset and generate view actions

Addition conditioning - anchor frame vs captions

🎯

Explore video generation results with anchor frames vs captions

JavisDiT audio-visual object editing

🎬

Add or remove objects in video with matching audio edits

Stage-2 Conditioning 4-Way

🧭

Generate future video and tactile frames from actions

JavisDiT vs AVI-Edit vs CoherentAVEdit vs InstructAV2AV

🎬

Explore and compare audio‑visual object add/remove models

Gel-Frame Action Encoder

🤚

Predict gel tactile flow from contact action and force

AVI-Edit vs JavisDiT vs InstructAV2AV

🎬

Compare video editing models across a full dataset

AEA Milestone Viewer

🎬

Egocentric video paired with Gemini milestone annotations

JavisDiT Edit — audiofix scratch 0811 step 46000 (test)

🔊

Explore video edit results with audio comparison

Force-aware visuo-tactile MoT

🤚

Explore force predictions from visuo‑tactile videos

JavisDiT Edit — srcconcat 0810 step 11000 (test)

🎬

Explore AI-generated video edits and compare results

JavisDiT source motion extraction

🌊

Explore extracted motion fields from video clips

Input / Output Pairs — AV Object-Removal Pipeline

🎬

Remove objects from videos and isolate their sounds