This paper presents a novel watermarking scheme designed to address the copyright protection challenges encountered with Neural radiation field(NeRF)models.We employ an embedding network to integrate the watermark int...This paper presents a novel watermarking scheme designed to address the copyright protection challenges encountered with Neural radiation field(NeRF)models.We employ an embedding network to integrate the watermark into the images within the training set.Then,theNeRFmodel is utilized for 3Dmodeling.For copyright verification,a secret image is generated by inputting a confidential viewpoint into NeRF.On this basis,design an extraction network to extract embedded watermark images fromconfidential viewpoints.In the event of suspicion regarding the unauthorized usage of NeRF in a black-box scenario,the verifier can extract the watermark from the confidential viewpoint to authenticate the model’s copyright.The experimental results demonstrate not only the production of visually appealing watermarks but also robust resistance against various types of noise attacks,thereby substantiating the effectiveness of our approach in safeguarding NeRF.展开更多
Academy of Sciences,Beijing 100029 2 Lanzhou University,Lanzhou 730000 3 Cold and Arid Regions Environmental and Engineering Research Institute, Chinese Academy of Sciences,Lanzhou 730000 4 Nanjing University of Infor...Academy of Sciences,Beijing 100029 2 Lanzhou University,Lanzhou 730000 3 Cold and Arid Regions Environmental and Engineering Research Institute, Chinese Academy of Sciences,Lanzhou 730000 4 Nanjing University of Information Science＆Technology,Nanjing 210044 Using both fast and slow electric field change sensors and field mill,multi-station observations on lightning flashes over China inland plateau areas were conducted during the summer of 2004.All of the stations were synchronized by GPS with a time-resolution of±50 ns.Using the different time of arrival（DTOA）and based on the fast electric field change sensor,a lightning radiation location technique was developed.Radiation pulses in the initial stages of five intracloud（IC）lightning discharges which occurred on 20 August were analyzed.The results indicate that the technique developed could effectively locate the lightning radiation sources.Furthermore,the lightning discharges were compared with the Doppler radar data.The results show that the radiation sources were well associated with the storm development.When the storm was at the mature stage with an echo top of about 10.9 km,the radiation sources ranged from 5.2 to 8.3 km above mean sea level;when the storm gradually became weaker,with echo top of about 7.9 km,the radiation sources ranged from 3.0 to 5.9 km.In particular,one of the IC lightning discharges ranged from 3.0 to 4.9 km during the dissipation stage of the storm.The results also indicate that the radiation sources were closely associated with the high reflectivity region（25–50 dBZ）of the storm,which,to some extent,demonstrates the reliability of the location results,thereby showing that multi-station observations of fast electric field change sensors could be a useful tool for monitoring the storm development.Location errors from radiation sources were also compared with the radar data and the results of a simple simulation.It was found that the errors were getting smaller when the radiation sources approached the center of the detection network, and vice versa.Compared with the limited experimental observations,the simulation results were found capable of effectively reflecting the location errors.展开更多
基金supported by the National Natural Science Foundation of China,with Fund Number 62272478.
文摘This paper presents a novel watermarking scheme designed to address the copyright protection challenges encountered with Neural radiation field(NeRF)models.We employ an embedding network to integrate the watermark into the images within the training set.Then,theNeRFmodel is utilized for 3Dmodeling.For copyright verification,a secret image is generated by inputting a confidential viewpoint into NeRF.On this basis,design an extraction network to extract embedded watermark images fromconfidential viewpoints.In the event of suspicion regarding the unauthorized usage of NeRF in a black-box scenario,the verifier can extract the watermark from the confidential viewpoint to authenticate the model’s copyright.The experimental results demonstrate not only the production of visually appealing watermarks but also robust resistance against various types of noise attacks,thereby substantiating the effectiveness of our approach in safeguarding NeRF.
基金the 100 Talents Program of the Chinese Academy of Sciencesthe Special Research Project for Public Welfare of China under Grant No.200704027-1the National Natural Science Foundation of China under Grant No.40930949
文摘Academy of Sciences,Beijing 100029 2 Lanzhou University,Lanzhou 730000 3 Cold and Arid Regions Environmental and Engineering Research Institute, Chinese Academy of Sciences,Lanzhou 730000 4 Nanjing University of Information Science＆Technology,Nanjing 210044 Using both fast and slow electric field change sensors and field mill,multi-station observations on lightning flashes over China inland plateau areas were conducted during the summer of 2004.All of the stations were synchronized by GPS with a time-resolution of±50 ns.Using the different time of arrival（DTOA）and based on the fast electric field change sensor,a lightning radiation location technique was developed.Radiation pulses in the initial stages of five intracloud（IC）lightning discharges which occurred on 20 August were analyzed.The results indicate that the technique developed could effectively locate the lightning radiation sources.Furthermore,the lightning discharges were compared with the Doppler radar data.The results show that the radiation sources were well associated with the storm development.When the storm was at the mature stage with an echo top of about 10.9 km,the radiation sources ranged from 5.2 to 8.3 km above mean sea level;when the storm gradually became weaker,with echo top of about 7.9 km,the radiation sources ranged from 3.0 to 5.9 km.In particular,one of the IC lightning discharges ranged from 3.0 to 4.9 km during the dissipation stage of the storm.The results also indicate that the radiation sources were closely associated with the high reflectivity region（25–50 dBZ）of the storm,which,to some extent,demonstrates the reliability of the location results,thereby showing that multi-station observations of fast electric field change sensors could be a useful tool for monitoring the storm development.Location errors from radiation sources were also compared with the radar data and the results of a simple simulation.It was found that the errors were getting smaller when the radiation sources approached the center of the detection network, and vice versa.Compared with the limited experimental observations,the simulation results were found capable of effectively reflecting the location errors.
