From b3295534b93d0be9a0548d4dd234e41c5e3c1f97 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?J=CF=83=CF=85=D0=B3=D0=98=D0=B5=CE=B3?= <83916236+journey-zhuang@users.noreply.github.com> Date: Wed, 4 Dec 2024 09:33:13 +0800 Subject: [PATCH 1/2] Update index.html --- index.html | 35 +++++++++++++++++++++++++++++++++++ 1 file changed, 35 insertions(+) diff --git a/index.html b/index.html index 879b03b28..0abd127b4 100644 --- a/index.html +++ b/index.html @@ -649,6 +649,41 @@

Ablation: Gaussian Refine

+
+
+ +
+
+

Related Work

+ +
+

+ Below are some related works that were introduced around the same time as ours. +

+

+ StyleGaussian: Instant 3D Style Transfer with Gaussian Splatting (Liu et al., 2024) introduces an approach for fast 3D style transfer using Gaussian splatting techniques. +

+

+ Instruct-GS2GS: Editing 3D Gaussian Splats with Instructions (Vachha & Haque, 2024) presents a framework for editing 3D Gaussian splats through natural language instructions. +

+

+ Styleshot: A Snapshot on Any Style (Gao et al., 2024) explores the ability to apply arbitrary styles to 3D objects using a snapshot-based approach. +

+

+ Ip-adapter: Text Compatible Image Prompt Adapter for Text-to-Image Diffusion Models (Ye et al., 2023) introduces an image prompt adapter for enhancing text-to-image models. +

+

+ Instantstyle: Free Lunch Towards Style-Preserving in Text-to-Image Generation (Wang et al., 2024) proposes a method for maintaining style consistency in text-to-image generation. +

+

+ 3D Gaussian Splatting for Real-Time Radiance Field Rendering (Kerbl et al., 2023) investigates real-time rendering techniques for 3D Gaussian splatting in radiance fields. +

+
+
+
+ +
+
From 4dcd8be80300edaa4ba21e5b8f8b6f1892d18209 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?J=CF=83=CF=85=D0=B3=D0=98=D0=B5=CE=B3?= <83916236+journey-zhuang@users.noreply.github.com> Date: Wed, 4 Dec 2024 09:33:58 +0800 Subject: [PATCH 2/2] Update index.html --- index.html | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/index.html b/index.html index 0abd127b4..3996a80f9 100644 --- a/index.html +++ b/index.html @@ -521,7 +521,7 @@

Qualitative Comparisons

- Qualitative Comparisons: We compare our method against other state-of-the-art (SOTA) approaches: StyleGaussian (Liu et al., 2024), IGS2GS (IGS2GS Reference), and Ref-NPR (Zhang et al., 2023) on three datasets (Chair, Hotdog, and Mic) using three styles: Cartoon, Sky Painting, and Fire. + Qualitative Comparisons: We compare our method against other state-of-the-art (SOTA) approaches: StyleGaussian (Liu et al., 2024), IGS2GS (IGS2GS Reference), and Ref-NPR (Zhang et al., 2023) on three datasets (Chair, Hotdog, and Mic) using three styles: Cartoon, Sky Painting, and Fire.

The horizontal axis represents the compared methods, while the vertical axis displays different datasets. Our method effectively preserves the original model details, including: