From 35bff53840a8ce3a7b7559396c8a7bb335f09e0f Mon Sep 17 00:00:00 2001 From: gianscarpe Date: Fri, 1 Mar 2024 10:55:38 +0100 Subject: [PATCH] updated readm --- README.md | 6 ++++++ _config.yaml | 2 ++ 2 files changed, 8 insertions(+) diff --git a/README.md b/README.md index 02c75c2..c825749 100644 --- a/README.md +++ b/README.md @@ -3,7 +3,13 @@ Reassembly tasks play a fundamental role in many fields and multiple approaches - [Code](https://github.com/IIT-PAVIS/DiffAssemble) - [Arxiv](soon) +# Method ![Method](https://raw.githubusercontent.com/IIT-PAVIS/DiffAssemble/release/page/DiffAssemble.png) +Following the Diffusion Probabilistic +Models formulations, we model a Markov chain where we inject noise into the pieces’ position and orientation. At timestep t = 0, the +pieces are in their correct position, and at timestep t = T , they are in a random position with random orientation. At each timestep t, our +attention-based GNN takes as input a graph where each node contains an equivariant feature that describes a particular piece and its position +and orientation. The network then predicts a less noisy version of the piece’s position and orientation. # Datasets diff --git a/_config.yaml b/_config.yaml index 8813df1..233de09 100644 --- a/_config.yaml +++ b/_config.yaml @@ -1,3 +1,5 @@ remote_theme: pages-themes/cayman@v0.2.0 plugins: - jekyll-remote-theme +title: DiffAssemble +description: Official repository for "DiffAssemble: A Unified Graph-Diffusion Model for 2D and 3D Reassembly" accepted at CVPR2024