IEEE ICIP 2025 || Anchorage, Alaska, USA || 14-18 September 2025

WE5.L3.2

ERPGS: EQUIRECTANGULAR IMAGE RENDERING ENHANCED WITH 3D GAUSSIAN REGULARIZATION

Shintaro Ito, Natsuki Takama, Koichi Ito, Tohoku University, Japan; Hwann-Tzong Chen, National Tsing Hua University, Japan; Takafumi Aoki, Tohoku University, Japan

Session:

WE5.L3: Image and Video Analysis, Synthesis, and Retrieval 4 Lecture

Location:

Room 1

Presentation Time:

Wed, 17 Sep, 16:45 - 17:00 Anchorage Time (UTC -8)

Session Chair:

Ling Guan, Toronto Metropolitan University and Ryerson University

Session WE5.L3

WE5.L3.1: SFS-NERF: ENHANCING GEOMETRY CONSISTENCY IN FEW-SHOT NOVEL VIEW SYNTHESIS THROUGH SURFACE-AWARE NEURAL RENDERING

Seonji Park, Nam Ik Cho, Seoul National University, Korea (South)

WE5.L3.2: ERPGS: EQUIRECTANGULAR IMAGE RENDERING ENHANCED WITH 3D GAUSSIAN REGULARIZATION

Shintaro Ito, Natsuki Takama, Koichi Ito, Tohoku University, Japan; Hwann-Tzong Chen, National Tsing Hua University, Japan; Takafumi Aoki, Tohoku University, Japan

WE5.L3.3: SIAVATAR: ANIMATABLE 3D GAUSSIAN AVATAR FROM A SINGLE IMAGE

Zonghao Lin, Ruiyan Wang, Jun Ling, Li Song, Shanghai Jiao Tong University, China

WE5.L3.4: IterDiff: Training-Free Iterative Face Editing via Efficient CLIP-guided Memory Bank

Chun-Yao Chiu, National Yang Ming Chiao Tung University, Taiwan; Feng-Kai Huang, National Taiwan University, Taiwan; Teng-Fang Hsiao, Hong-Han Shuai, National Yang Ming Chiao Tung University, Taiwan; Wen-Huang Cheng, National Taiwan University, Taiwan

WE5.L3.5: STABLE-INVERTIBLE GRAPH CONVOLUTIONAL NETWORKS FOR LABEL-EFFICIENT SKELETON-BASED RECOGNITION

Hichem Sahbi, CNRS Sorbonne University, France

WE5.L3.6: DIVERSIFYING HUMAN POSE IN SYNTHETIC DATA FOR AERIAL-VIEW HUMAN DETECTION

Yi-Ting Shen, University of Maryland, College Park, United States; Hyungtae Lee, BlueHalo, United States; Heesung Kwon, DEVCOM Army Research Laboratory, United States; Shuvra S. Bhattacharyya, University of Maryland, College Park, United States