In temporal action localization methods, temporal downsampling operations are widely used to extract proposal features, but they often lead to the aliasing problem, due to lacking consideration of sampling rates. This paper aims to verify the existence of aliasing in TAL methods and investigate utilizing low pass filters to solve this problem by inhibiting the high-frequency band. However, the high-frequency band usually contains large amounts of specific information, which is important for model inference. Therefore, it is necessary to make a tradeoff between anti-aliasing and reserving high-frequency information. To acquire optimal performance, this paper learns different cutoff frequencies for different instances dynamically. This design can be plugged into most existing temporal modeling programs requiring only one additional cutoff frequency parameter. Integrating low pass filters to the downsampling operations significantly improves the detection performance and achieves comparable results on THUMOS'14, ActivityNet~1.3, and Charades datasets. Experiments demonstrate that anti-aliasing with low pass filters in TAL is advantageous and efficient.
Revisiting Light Field Rendering with Deep Anti-Aliasing Neural Network
Northeastern University（东北大学）；Institute of Industrial Artificial Intelligence, Northeastern University（东北大学工业人工智能研究院）；Tsinghua University（清华大学）；Beijing National Research Center for Information Science and Technology（北京信息科学与技术国家研究中心）
SA-GS: Scale-Adaptive Gaussian Splatting for Training-Free Anti-Aliasing
Tongji University（同济大学）；Ocean University of China（中国海洋大学）；Institute for AI Industry Research (AIR), Tsinghua University（清华大学人工智能产业研究院）；Duke Kunshan University（昆山杜克大学）
PrismGS: Physically-Grounded Anti-Aliasing for High-Fidelity Large-Scale 3D Gaussian Splatting
School of Information Science and Electronic Engineering, Shanghai Jiao Tong University, Shanghai, China（上海交通大学信息科学与电子工程学院）；Cooperative Mediant Innovation Center, Shanghai Jiao Tong University, Shanghai, China（上海交通大学协同创新中心）；College of Information Science and Technology, Eastern Institute of Technology, Ningbo, China（宁波工程技术学院信息科学与技术学院）
PrismGS提出物理基础正则化框架，通过金字塔多尺度监督和尺寸下界约束，解决大规模3DGS在高分辨率下的锯齿与闪烁问题，实现即插即用，在多个城市数据集上较CityGaussian提升约1.5 dB PSNR。
Rip-NeRF: Anti-aliasing Radiance Fields with Ripmap-Encoded Platonic Solids
Beihang University（北京航空航天大学）；Tencent AI Lab（腾讯人工智能实验室）；Beijing Institute of Technology（北京理工大学）；Tsinghua University（清华大学）；Dermatology Hospital（皮肤病医院）；Southern Medical University（南方医科大学）
State Key Laboratory of Robotics and Systems (HIT), Harbin Institute of Technology（机器人系统国家重点实验室（哈工大））；Yangtze River Delta HlT Robot Technology Research Institute（长江三角洲HLT机器人技术研究院）；Department of Computer Science, National University of Singapore（新加坡国立大学计算机科学系）
