NEW CC Perspectives on Levels of Conspiracy & Corruption in 2020 Election, CV19 & DS Existence

  • Christian Chat is a moderated online Christian community allowing Christians around the world to fellowship with each other in real time chat via webcam, voice, and text, with the Christian Chat app. You can also start or participate in a Bible-based discussion here in the Christian Chat Forums, where members can also share with each other their own videos, pictures, or favorite Christian music.

    If you are a Christian and need encouragement and fellowship, we're here for you! If you are not a Christian but interested in knowing more about Jesus our Lord, you're also welcome! Want to know what the Bible says, and how you can apply it to your life? Join us!

    To make new Christian friends now around the world, click here to join Christian Chat.

Was 2020 US Election Fraudulent AND was CV19 a Deep State Plandemic?


  • Total voters
    109

ZNP

Well-known member
Sep 14, 2020
36,851
6,755
113
It's hard to tell!
At this point
1) the DS is so desperate that they will pull out every Dirty Trick in their Diabolic Playbook to prevent a Fair 2024 Election as their ONLY means to maintain their RUSE of DECEIT and their last hope to RESUME their GLOBALIST AGENDA...
2) the White Hats are willing to let them EXPOSE THEMSELVES for all the WORLD to SEE their DIABOLIC PLANs
3) I am convinced that the DS will succeed in creating CHAOS and MAYHEM in the streets of the USA as a means to instigate a CIVIL WAR and to KEEP the American PEOPLE DIVIDED...
4) Like any successful STING OPERATION that intends to CUT OFF the HEADS of the EVIL HYDRA - the Major FEAR EVENT has to be completely BELIEVABLE to EVERYONE and the EXPOSURE has to be BLATANT for the WORLD to SEE!!!

So far, all their PROXY WARS Russia-Ukraine and Israel-Hamas have played a ROLE in their efforts to maintain their RUSE of DECEIT...
DS Diabolic Moves with China and/or Iran Remain on the TABLE, as is an ALIEN INVASION...
Unfortunately, because we remain in the MIDST of an INFORMATION WARFARE - we remain vulnerable to dis-information and mis-information when it comes to these potential Proxy War Moves and accuracy of immediate impacts, implications and effects...

At this point my greatest FEAR is a CIVIL WAR in the streets of the USA... Given the growing numbers of Young Male Adults crossing our Borders - foot soldiers for the DS...

Feeling a bit more comfortable (further distant) that the Global White Hat ALLIANCE is completely capable of addressing those potential near- mid- and long-term World Warfare Operations...

Bottomline - the DS will likely instigate CYBER WARFARE... and we are TOLD that the White Hats have COUNTER MOVES and MEASURES in place for all of the DS remaining Moves... Star-link will be KEY for the White Hats to PUSH TRUTH to the PEOPLE when the DS attempts to turn out the lights...

Checkmate has already been called between the Forces of Good vs Evil... This is why the DS moves are already known and anticipated... as they are in a position of COMPLETE DESPERATION.
Hard to tell? It's hard to tell if WW3 has already started, that is what is hard to tell.
 

Solemateleft

Honor, Courage, Commitment
Jun 25, 2017
13,871
4,033
113
Hard to tell? It's hard to tell if WW3 has already started, that is what is hard to tell.
Yes... We most certainly are in a NEW KINDA WWIII given the nature of this IRREGULAR, INFORMATION 5G Warfare....
I interpreted his Question in the CONTEXT of a FULL BLOWN PEARL HARBER 2.0 - w/NO DOUBT the USA is/will be LAUNCHING FORCES into KINETIC COMBAT & Warfare...
 

ZNP

Well-known member
Sep 14, 2020
36,851
6,755
113
Yes... We most certainly are in a NEW KINDA WWIII given the nature of this IRREGULAR, INFORMATION 5G Warfare....
I interpreted his Question in the CONTEXT of a FULL BLOWN PEARL HARBER 2.0 - w/NO DOUBT the USA is/will be LAUNCHING FORCES into KINETIC COMBAT & Warfare...
The US has spent so much money to automate our army with drones, robots, and AI it is very possible that we will be fighting a full blown war before they even have a draft.
 

Solemateleft

Honor, Courage, Commitment
Jun 25, 2017
13,871
4,033
113

Solemateleft

Honor, Courage, Commitment
Jun 25, 2017
13,871
4,033
113
X22 Report Live: The Stage Must Be Set! Crumbs Are Easy To Swallow! America Is No Longer For Sale! - Must Video
Wednesday, January 3, 2024 20:32

The [DS] is now panicking, the Epstein is coming out, crumbs are easy to swallow, this will go slow then speed up. The stage must be set. The [DS] has infiltrated the country and have been selling America secrets to the highest bidder. America is no longer for sale, the patriots have operations that are active and the [DS] power is removed. This is about showing the American people the infiltration and having the people take the country back.
 

ZNP

Well-known member
Sep 14, 2020
36,851
6,755
113

Solemateleft

Honor, Courage, Commitment
Jun 25, 2017
13,871
4,033
113
Do you know what a kill bot is, these very small drones that identify a face and then hit them with a targeted explosion that kills them. Unleash a million of them on a city and see how many people survive hiding in boxes.
LOL, you've been watching too much SCI-FI...
Yes, we have and are exploring with LOTS of HIGH TECH capabilities... I work in the Capability Development business for the USMC that includes AI... The ISSUE with this exploratory Tech as the Marines have PROVEN - is that it is NOT PERFECT, it has NOT been Validated...
Asset Recognition is FLAWED... A Cherry Picker can easily be mistaken for a Rocket Launcher...
Cloned Masks for DOUBLES or Disguises are very easy defense against Facial Recognition...

The PROBLEM with AI in DoD is DATA... BAD DATA = BAD UNTRUSTWORTHY AI...

The Department of Defense’s Chief Digital and Artificial Intelligence Office (CDAO) is the senior official responsible for the acceleration of the DoD’s adoption of data, analytics, and AI to generate decision advantage across, from the boardroom to the battlefield.
  • Review and more tightly integrate the Department’s policy, strategy, and governance of data, analytics, AI, to include an integrated Data, Analytics and AI Strategy as well as maturing a Responsible AI Ecosystem.
  • We are MANNDATING Authoritative Data and Trustworthy AI.... We are a VERY LONG WAY away from these REALITIES...
U.S. DEPARTMENT OF DEFENSE RESPONSIBLE ARTIFICIAL INTELLIGENCE STRATEGY AND IMPLEMENTATION PATHWAY
 

ZNP

Well-known member
Sep 14, 2020
36,851
6,755
113
LOL, you've been watching too much SCI-FI...
Yes, we have and are exploring with LOTS of HIGH TECH capabilities... I work in the Capability Development business for the USMC that includes AI... The ISSUE with this exploratory Tech as the Marines have PROVEN - is that it is NOT PERFECT, it has NOT been Validated...
Asset Recognition is FLAWED... A Cherry Picker can easily be mistaken for a Rocket Launcher...
Cloned Masks for DOUBLES or Disguises are very easy defense against Facial Recognition...

The PROBLEM with AI in DoD is DATA... BAD DATA = BAD UNTRUSTWORTHY AI...

The Department of Defense’s Chief Digital and Artificial Intelligence Office (CDAO) is the senior official responsible for the acceleration of the DoD’s adoption of data, analytics, and AI to generate decision advantage across, from the boardroom to the battlefield.
  • Review and more tightly integrate the Department’s policy, strategy, and governance of data, analytics, AI, to include an integrated Data, Analytics and AI Strategy as well as maturing a Responsible AI Ecosystem.
  • We are MANNDATING Authoritative Data and Trustworthy AI.... We are a VERY LONG WAY away from these REALITIES...
U.S. DEPARTMENT OF DEFENSE RESPONSIBLE ARTIFICIAL INTELLIGENCE STRATEGY AND IMPLEMENTATION PATHWAY
The problem I have with all this is that we are frequently told that the military is 25 years ahead of what we see in the public sector. We know that the Patriot act was a massive retooling of the US military because Dick Cheney had said a year earlier he needed to make these changes. That tells me in 2000 they had the drone technology and they have developed it over the last 23 years. All those many hours of people flying drones gives data for AI to be trained to fly them themselves. Also, all this facial recognition software has been training, if the public sector is already very good at that I have to believe the military is way ahead.

So you can see my cognitive dissonance. If you take what they show us in public and project out 20-25 years it will be amazing tech, so that is what I assume the military has today, though you would need a top secret clearance to see that.
 

Solemateleft

Honor, Courage, Commitment
Jun 25, 2017
13,871
4,033
113
The problem I have with all this is that we are frequently told that the military is 25 years ahead of what we see in the public sector. We know that the Patriot act was a massive retooling of the US military because Dick Cheney had said a year earlier he needed to make these changes. That tells me in 2000 they had the drone technology and they have developed it over the last 23 years. All those many hours of people flying drones gives data for AI to be trained to fly them themselves. Also, all this facial recognition software has been training, if the public sector is already very good at that I have to believe the military is way ahead.

So you can see my cognitive dissonance. If you take what they show us in public and project out 20-25 years it will be amazing tech, so that is what I assume the military has today, though you would need a top secret clearance to see that.
NOPE!!! the REALITY is quite the OPPOSITE...
DOD simply DOES NOT have the BUDGET and funding and access to Trustworthy Data as Industry...
Granted there are lots of Top Secret stuff that DARPA plays around with - (lol, much of the origins of the Deep State Black Hat White Hat origins are assumed to stem from an AI project.) Black Op Funds is another story... but NOT a REALITY in practice...
BUT, in practice - DOD is WAY BEHIND Industry when it comes to the use of AI in NONE WARFARE applications...
ie FOLLOW the DATA... Industry has Data in Banking, Automotive, Personnel/HR, etc... Unfortunately - these types of DATA do NOT translate into Military Applications very well..
 

ZNP

Well-known member
Sep 14, 2020
36,851
6,755
113
NOPE!!! the REALITY is quite the OPPOSITE...
DOD simply DOES NOT have the BUDGET and funding and access to Trustworthy Data as Industry...
Granted there are lots of Top Secret stuff that DARPA plays around with - (lol, much of the origins of the Deep State Black Hat White Hat origins are assumed to stem from an AI project.) Black Op Funds is another story... but NOT a REALITY in practice...
BUT, in practice - DOD is WAY BEHIND Industry when it comes to the use of AI in NONE WARFARE applications...
ie FOLLOW the DATA... Industry has Data in Banking, Automotive, Personnel/HR, etc... Unfortunately - these types of DATA do NOT translate into Military Applications very well..
Well that is not what some of my sources say, no way to verify what they say, but they come across as very believable. We all know that the gold plated toilets were not gold plated toilets but simply a way to hide secret programs. We know the amount of money the US spends on the military and secret budgets is astounding by any measure and far more than any other country, we know that they say one thing about not controlling our media but that is all a lie, they completely controlled Facebook, Twitter, etc.

So I don't believe it, they decide that Facebook is the winner, Myspace is the loser, that tells me that Zuckerburg is their puppet. We know that Twitter also did their bidding. No reason to think Google wasn't another one of the companies they controlled. Not long ago we discovered they were fully into AT&T so that they could listen to every call.

If they control these companies I don't believe they would let them develop and release AI that is considered more dangerous than an atomic bomb if they weren't a couple of decades ahead of them. AI requires "big data" and who has more data than the NSA?
 

Solemateleft

Honor, Courage, Commitment
Jun 25, 2017
13,871
4,033
113
The problem I have with all this is that we are frequently told that the military is 25 years ahead of what we see in the public sector. We know that the Patriot act was a massive retooling of the US military because Dick Cheney had said a year earlier he needed to make these changes. That tells me in 2000 they had the drone technology and they have developed it over the last 23 years. All those many hours of people flying drones gives data for AI to be trained to fly them themselves. Also, all this facial recognition software has been training, if the public sector is already very good at that I have to believe the military is way ahead.

So you can see my cognitive dissonance. If you take what they show us in public and project out 20-25 years it will be amazing tech, so that is what I assume the military has today, though you would need a top secret clearance to see that.
Yes, there are lots of advances with Drones... but DoD at large is a long way away from full autonomy of robotics... The USMC literally just stood up Robotics as a capability portfolio this past year... still just a Concept... and the USMC simply does NOT have the budget to afford to implement all of their AI visions...
 

ZNP

Well-known member
Sep 14, 2020
36,851
6,755
113
Yes, there are lots of advances with Drones... but DoD at large is a long way away from full autonomy of robotics... The USMC literally just stood up Robotics as a capability portfolio this past year... still just a Concept... and the USMC simply does NOT have the budget to afford to implement all of their AI visions...
Then why the debate at the UN about allowing AI to target and kill individuals? Sounds to me like they are ready to do that right now.

Disinformation is part of warfare. Making people believe there is nothing to worry about stupid AI can be a simple head fake.
 

Solemateleft

Honor, Courage, Commitment
Jun 25, 2017
13,871
4,033
113
Well that is not what some of my sources say, no way to verify what they say, but they come across as very believable. We all know that the gold plated toilets were not gold plated toilets but simply a way to hide secret programs. We know the amount of money the US spends on the military and secret budgets is astounding by any measure and far more than any other country, we know that they say one thing about not controlling our media but that is all a lie, they completely controlled Facebook, Twitter, etc.

So I don't believe it, they decide that Facebook is the winner, Myspace is the loser, that tells me that Zuckerburg is their puppet. We know that Twitter also did their bidding. No reason to think Google wasn't another one of the companies they controlled. Not long ago we discovered they were fully into AT&T so that they could listen to every call.

If they control these companies I don't believe they would let them develop and release AI that is considered more dangerous than an atomic bomb if they weren't a couple of decades ahead of them. AI requires "big data" and who has more data than the NSA?
I led a study on Big Data Analytics for the USMC a few years ago... so I'll just leave it at that...
 

Solemateleft

Honor, Courage, Commitment
Jun 25, 2017
13,871
4,033
113
Then why the debate at the UN about allowing AI to target and kill individuals? Sounds to me like they are ready to do that right now.
Yes... we certainly have a Drone capability... but like I said... We have a problem with Responsible & Trustworthy AI...
 

Solemateleft

Honor, Courage, Commitment
Jun 25, 2017
13,871
4,033
113
Then why the debate at the UN about allowing AI to target and kill individuals? Sounds to me like they are ready to do that right now.
Disinformation is part of warfare. Making people believe there is nothing to worry about stupid AI can be a simple head fake.
If DoD already has the BROADER AI capability - then why do they have a NEW Data, Analytics and AI Adoption Strategy?
chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://media.defense.gov/2023/Nov/02/2003333300/-1/-1/1/DOD_DATA_ANALYTICS_AI_ADOPTION_STRATEGY.PDF
chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://media.defense.gov/2023/Nov/02/2003333300/-1/-1/1/DOD_DATA_ANALYTICS_AI_ADOPTION_STRATEGY.PDF
1704335604629.png
 

ZNP

Well-known member
Sep 14, 2020
36,851
6,755
113
Ongoing Face Recognition Vendor Test (FRVT) Part 2: Identification

This study was published in 2018.

Executive Summary This report documents performance of face recognition algorithms submitted for evaluation on image datasets maintained at NIST. The algorithms implement one-to-many identification of faces appearing in two-dimensional images. The primary dataset is comprised of 26.6 million reasonably well-controlled live portrait photos of 12.3 million individuals. Three smaller datasets containing more unconstrained photos are also used: 3.2 million webcam images; 2.5 million photojournalism and amateur photographer photos; and 90 thousand faces cropped from surveillance-style video clips. The report will be useful for comparison of face recognition algorithms, and assessment of absolute capability. The report details recognition accuracy for 127 algorithms from 45 developers, associating performance with participant names. The algorithms are prototypes, submitted in February and June 2018 by research and development laboratories of commercial face recognition suppliers and one university. The algorithms were submitted to NIST as compiled libraries and are evaluated as black boxes behind a NIST-specified C++ testing interface. The report therefore does not describe how algorithms operate. The evaluation was run in two phases, starting Feburary and June 2018 respectively, with developers receiving technical feedback after each. A third phase commenced on October 30, 2018, results from which will reported in the first quarter of 2019. The major result of the evaluation is that massive gains in accuracy have been achieved in the last five years (2013- 2018) and these far exceed improvements made in the prior period (2010-2013). While the industry gains are broad - at least 28 developers’ algorithms now outperform the most accurate algorithm from late 2013 - there remains a wide range of capabilities. With good quality portrait photos, the most accurate algorithms will find matching entries, when present, in galleries containing 12 million individuals, with error rates below 0.2%. The remaining errors are in large part attributable to long-run ageing and injury. However, for at least 10% of images - those with significant ageing or sub-standard quality - identification often succeeds but recognition confidence is diminished such that matches become indistinguishable from false positives, and human adjudication becomes necessary. The accuracy gains stem from the integration, or complete replacement, of prior approaches with those based on deep convolutional neural networks. As such, face recognition has undergone an industrial revolution, with algorithms increasingly tolerant of poor quality images. Whether the revolution continues or has moved into a more evolutionary phase, further gains can be expected as machine learning architectures further develop, larger datasets are assembled and benchmarks are further utilized.


https://nvlpubs.nist.gov/nistpubs/ir/2018/NIST.IR.8238.pdf

It is safe to assume that their ability has improved, most would agree that 99.8% is a very high success rate. The main issue they have is identifying people from old photographs of poor quality.

My point is that AI can identify a face accurately. If the military put many of these faces on a kill list AI can identify them and let the military know when they have spotted them, or if AI is allowed to make the kill just go ahead and kill them after identifying them.

Clearview AI is already being used by the US government to identify faces for a variety of security threats.
 

Solemateleft

Honor, Courage, Commitment
Jun 25, 2017
13,871
4,033
113
Well that is not what some of my sources say, no way to verify what they say, but they come across as very believable. We all know that the gold plated toilets were not gold plated toilets but simply a way to hide secret programs. We know the amount of money the US spends on the military and secret budgets is astounding by any measure and far more than any other country, we know that they say one thing about not controlling our media but that is all a lie, they completely controlled Facebook, Twitter, etc.

So I don't believe it, they decide that Facebook is the winner, Myspace is the loser, that tells me that Zuckerburg is their puppet. We know that Twitter also did their bidding. No reason to think Google wasn't another one of the companies they controlled. Not long ago we discovered they were fully into AT&T so that they could listen to every call.

If they control these companies I don't believe they would let them develop and release AI that is considered more dangerous than an atomic bomb if they weren't a couple of decades ahead of them. AI requires "big data" and who has more data than the NSA?
At the same time You are correct... in that DoD is a bit of a conundrum... Given the Corrupt CIA-DARPA Black Hat AI Projects...
Evidently those of us in the CHEAP seats only have visibility of the actual AI projects that have been FIELDED on the up and up...
PROJECT “LOOKING GLASS”, THE PLANS OF THE SECRET GOVERNMENT ON TIME MACHINES AND STARGATE DEVICES
AWAKENING, DISCLOSURE, POLITICS by PARADIGMSHYFT13JULY 18, 2021
 

ZNP

Well-known member
Sep 14, 2020
36,851
6,755
113
"The Chief Digital and Artificial Intelligence Office" sounds like the military is well passed the research and development stage.
 

ZNP

Well-known member
Sep 14, 2020
36,851
6,755
113
At the same time You are correct... in that DoD is a bit of a conundrum... Given the Corrupt CIA-DARPA Black Hat AI Projects...
Evidently those of us in the CHEAP seats only have visibility of the actual AI projects that have been FIELDED on the up and up...
PROJECT “LOOKING GLASS”, THE PLANS OF THE SECRET GOVERNMENT ON TIME MACHINES AND STARGATE DEVICES
AWAKENING, DISCLOSURE, POLITICS by PARADIGMSHYFT13JULY 18, 2021
NSA would be at the cutting edge, US marines would be the very last to have this capability.
 

Solemateleft

Honor, Courage, Commitment
Jun 25, 2017
13,871
4,033
113
Ongoing Face Recognition Vendor Test (FRVT) Part 2: Identification

This study was published in 2018.

Executive Summary This report documents performance of face recognition algorithms submitted for evaluation on image datasets maintained at NIST. The algorithms implement one-to-many identification of faces appearing in two-dimensional images. The primary dataset is comprised of 26.6 million reasonably well-controlled live portrait photos of 12.3 million individuals. Three smaller datasets containing more unconstrained photos are also used: 3.2 million webcam images; 2.5 million photojournalism and amateur photographer photos; and 90 thousand faces cropped from surveillance-style video clips. The report will be useful for comparison of face recognition algorithms, and assessment of absolute capability. The report details recognition accuracy for 127 algorithms from 45 developers, associating performance with participant names. The algorithms are prototypes, submitted in February and June 2018 by research and development laboratories of commercial face recognition suppliers and one university. The algorithms were submitted to NIST as compiled libraries and are evaluated as black boxes behind a NIST-specified C++ testing interface. The report therefore does not describe how algorithms operate. The evaluation was run in two phases, starting Feburary and June 2018 respectively, with developers receiving technical feedback after each. A third phase commenced on October 30, 2018, results from which will reported in the first quarter of 2019. The major result of the evaluation is that massive gains in accuracy have been achieved in the last five years (2013- 2018) and these far exceed improvements made in the prior period (2010-2013). While the industry gains are broad - at least 28 developers’ algorithms now outperform the most accurate algorithm from late 2013 - there remains a wide range of capabilities. With good quality portrait photos, the most accurate algorithms will find matching entries, when present, in galleries containing 12 million individuals, with error rates below 0.2%. The remaining errors are in large part attributable to long-run ageing and injury. However, for at least 10% of images - those with significant ageing or sub-standard quality - identification often succeeds but recognition confidence is diminished such that matches become indistinguishable from false positives, and human adjudication becomes necessary. The accuracy gains stem from the integration, or complete replacement, of prior approaches with those based on deep convolutional neural networks. As such, face recognition has undergone an industrial revolution, with algorithms increasingly tolerant of poor quality images. Whether the revolution continues or has moved into a more evolutionary phase, further gains can be expected as machine learning architectures further develop, larger datasets are assembled and benchmarks are further utilized.

https://nvlpubs.nist.gov/nistpubs/ir/2018/NIST.IR.8238.pdf

It is safe to assume that their ability has improved, most would agree that 99.8% is a very high success rate. The main issue they have is identifying people from old photographs of poor quality.

My point is that AI can identify a face accurately. If the military put many of these faces on a kill list AI can identify them and let the military know when they have spotted them, or if AI is allowed to make the kill just go ahead and kill them after identifying them.

Clearview AI is already being used by the US government to identify faces for a variety of security threats.
That is very impressive insight and research... No doubt... ACCESS to GOOD Trustworthy DATA is KEY...

BUT the translation into MILITARY APPLICATION is NOT trivial when a HUMAN is allowed the opportunity to TRICK the AI with MASKS... For example What if a stadium of students wore the same FACE MASK as your intended Target as DECOYS?
  1. www.wired.com › story › foundations-ai-riddled-errorsThe Foundations of AI Are Riddled With Errors | WIRED

    Mar 31, 2021 · The Foundations of AI Are Riddled With Errors | WIRED Will Knight Business Mar 31, 2021 7:00 AM The Foundations of AI Are Riddled With Errors The labels attached to images used to train...
  2. tech.co › news › list-ai-failures-mistakes-errorsAI Gone Wrong: A List of AI Errors, Mistakes and Failures

    2 days ago · AI Gone Wrong: An Updated List of AI Errors, Mistakes and Failures 2023 We track the latest AI errors and mistakes from 2023 and previous years, and catalog AI's worst failures to
  3. www.technologyreview.com › 2021/04/01 › 1021619AI datasets are filled with errors. It's warping what we know ...

    Apr 1, 2021 · April 1, 2021 Jeremy Lwanga/Unsplash The 10 most cited AI data sets are riddled with label errors, according to a new study out of MIT, and it’s distorting our understanding of the field’s...
    • Author: Karen Hao
  4. neurosciencenews.com › visual-recognition-ai-25229Tool Reveals Neural Network Errors in Image Recognition

    Nov 16, 2023 · The tool offers a visual representation of how the network categorizes images, highlighting areas of confusion. By mapping relationships between images, researchers can identify instances where the network struggles to differentiate between classifications, improving transparency in AI decision-making.
  5. www.csail.mit.edu › news › image-recognitionImage recognition accuracy: An unseen challenge confounding ...

    Dec 15, 2023 · MIT Computer Science and Artificial Intelligence Laboratory (CSAIL) researchers were surprised to find that despite the critical importance of understanding visual data in pivotal areas ranging from health care to transportation to household devices, the notion of an image’s recognition difficulty for humans has been almost entirely ignored.
  6. www.microsoft.com › en-us › researchImage Recognition: Current Challenges and Emerging ...

    Nov 1, 2018 · For critical applications such as autonomous driving, the cost of recognition errors is very high. So enormous datasets, containing hundreds of millions of images with rich annotations,