Noticias de León y provincia. Toda la información de actualidad, sucesos, política, deportes, cultura y Bierzo. Diario digital

Occidente ‘woke’

WOKEN is a web-based career exploration platform that educates, motivates \u0026 arms young professionals with the knowledge, tools \u0026 resources to confidently identify a fulfilling career path

WOKEN

Hey Hackers, WOKEN has been nominated in HackerNoon's annual Startup of the Year awards in New York City, NY, US. Please vote for us here: https://hackernoon.com/startups/north-america/north-america-manhattan-nyc-ny Read more about us below to understand why we deserve your vote. Are you also participating in SOTY 2023? If so, click here to fill out this interview. Meet WOKEN WOKEN is a career coaching company that offers software and coaching sessions to ensure that you can be intentional with every career decision you make. We guide professionals to confidently narrow in on their best fit, most suitable career path, support your upskilling decisions, strengthen your personal branding materials, and guide you through an intentional job search to land seamlessly in your next role. My Role I’m Rachel Serwetz, the CEO, Founder, and Head Career Exploration Coach at WOKEN. I started out in the corporate world (Goldman Sachs, Bridgewater Associates) and for the past 5.5 years, I have been building and bootstrapping WOKEN. I have a Tech MBA from NYU Stern and I am an ICF-Certified (PCC level) coach. How We're Disrupting/Improving the Career Industry We’ve never had an efficient, practical, effective means to clarify our best-fit career path. We’ve never known of the concept, nor the structure or tools, to have a pre-job search phase, something we call “career path exploration.” Whether you’re doing a small amount of reflection or making a big change, it’s critical to think about what you want before you leap into job applications. Career path exploration is a novel, step-by-step process of learning and reflection so you can compare, contrast, prioritize, and confidently identify which role, industry, and environment is the best fit for you. Standing Out from The Crowd Most other job tools focus on job applications and simply want you to land any job quickly; we focus on helping you strategically, and intentionally job search to find the right path, and not just the next thing. Most other career tools don’t follow a process, they have a ton of resources thrown at you and you don’t know where to start; for us, it’s a one-step-at-a-time process. Most other career tools don’t have accountability + a workspace combined; we provide guidance, resources, tools, coaching, and a workspace all in one. We are also holistic - ensuring you consider your mindset, accountability, and practical strategy, all at once. Most other coaching services are insanely expensive; our prices have several accessible tiers, ranging between $49 and $399 per month. And more! Our Predictions/Thoughts on the Career Industry in 2023 has proven and shown us that people are ready to commit and invest in themselves, that they want more from work, and that they want to find their purpose. We’re here to deliver an accessible, affordable means to navigate and reach that goal. The Great Resignation There’s been a ton of layoffs, which means that people now more than ever, need processes and tools to run an efficient but also effective job search. There’s also been a boom in AI. With an ever-changing world, fast-evolving tech, and a ton of new information, the stress around career choices simply grows. We’re here to organize your process so you can have support in navigating what’s out there. So you can learn both about yourself and your career options. What word defines the state of Careers in 2023? Stress / Mayhem. Managers and employees are burnout. No one wants to go back into the office. People are laid off/unemployed, and others are quitting. Statistically, a majority of people are disengaged at work or in jobs that are misaligned with their personality. This affects our well-being, our relationships, our wealth potential, our productivity and innovation, and our personal and professional lives, and in turn, it affects our companies and societies at large. Why we decided to participate in HackerNoon's Startup of the Year awards We want to remind the world that you can and should find a job you love. We want every professional to feel supported at every career step in their journey. Final Thoughts We are here to help and are committed to this mission! Please follow, get in touch, and we’d love to help with your career growth and advancement. Woken is nominated as ! Startup of the Year in Manhattan, NYC. Vote for us today

Helping Build Sustainable Careers - Interview with Startups of the Year Nominee, Woken 

Ayudando a construir carreras sostenibles: entrevista con el nominado a Startups del año, Woken

Aider à bâtir des carrières durables - Entretien avec le nominé des startups de l'année, Woken

टिकाऊ करियर बनाने में मदद करना - वर्ष के नामांकित व्यक्ति, वोकन के स्टार्टअप के साथ साक्षात्कार

持続可能なキャリア構築を支援 - Startups of the Year ノミネート、Woken へのインタビュー

Ajudando a construir carreiras sustentáveis - Entrevista com o candidato a startups do ano, Woken

Giúp Xây dựng Sự nghiệp Bền vững - Phỏng vấn Ứng cử viên Khởi nghiệp của Năm, Waken

帮助建立可持续的职业生涯 - 采访年度初创公司提名者 Woken

This paper is available on arxiv under CC BY-NC-SA 4.0 DEED license. Authors: (1) Alyvia Walters, Rutgers University, USA; (2) Tawfiq Ammaris, Rutgers University, USA; (3) Kiran Garimella, Rutgers University, USA; (4) Shagun Jhaver, Rutgers University, USA. Table of Links Abstract & Introduction Background and Related Work Methods The Tools of CRT Meme Production Knowledge Production in a Post-Truth World Conclusion & References The Tools of CRT Meme Production After iteratively coding each meme, we were left with 21 unique codes which could be subsumed under one or more of the following major rhetorical tactics: 1) struggles over definition, or how the meme makes sense of what critical race theory is; 2) constructing “antiracism,” or the ways in which the meme’s ideologies are coded as definitively not-racist to its intended audience; and 3) appeals to authority, or the ways in which the meme uses people or symbols to appear correct. These strategies, then, appear to be the most salient paths through which both pro- and anti-CRT arguments within these memes are built. Defining CRT With two exceptions, neither the pro- nor anti-CRT memes analyzed appear to be concerned with the “real” critical legal studies definition of critical race theory. Because CRT was not generally circulated within public discourse prior to the early 2020s, as mentioned above, there was a wide berth for political and epistemological work to be done in the construction of this definition in the public consciousness. Our analysis reveals that these memes do just that: while both proand anti-CRT memes provide a technically incorrect definition of what CRT actually is, the politics of sensemaking unfolds within these memes. Their consumers are left with bifurcating definitions of the bounds, risks, and benefits of critical race theory that ultimately serve to re-define the bounds and values of the communities in which these memes circulate. Within pro-CRT memes, the aggregated definition reads something like this: critical race theory means teaching history accurately and not being a racist and/or a Republican. These memes went to far fewer lengths than anti-CRT memes to define what CRT actually is, and relied instead upon defining it against other things: racism, Republicans, and/or the erasure of history. The top-circulated meme in our analysis is a prime example of this. In this meme, artist Jonathan Harris stands alongside his now-viral artwork entitled “Critical Race Theory,” which depicts the literal whitewashing of Black history (Figure 1). While this does nothing to define critical race theory, it certainly defines what it is not: the erasure of America’s violent, racist past. Similarly, image cluster 4 (Figure 2) is a computer-generated text-heavy meme that reads, “Republicans are not afraid of critical race theory. They don’t even know what it is. They’re afraid of theories critical of racists. They know who they are.” The irony, of course, is that this meme also does not indicate a real definition of CRT, or an indication of “knowing what it is”–it simply defines CRT against racists and Republicans, both of which believers in CRT cannot be. In contrast to this strategy, anti-CRT memes often utilize quite specific points of definition. Take, for instance, cluster 22 (Figure 3): In providing a six-point bulleted list, this meme lays out, in no uncertain terms, how its consumers are meant to understand CRT. These definitional strategies, however, are not always so concrete as a bulleted list. Whereas proCRT memes defined CRT against other ideas, anti-CRT memes sometimes worked to define it by conflating it with other “anti-American” ideas, such as Marxism (Figure 4) and straying from Christianity (Figure 5). In other words, antiCRT memes often tied CRT to other “woke” ideologies in order to define it, even as “wokeness,” in and of itself, lacks definitive boundaries. In all, anti-CRT memes essentially define critical race theory in the following way: CRT is a racist idea that makes people believe that race matters more than it should, and it is yet another way that “wokeness” is destroying America. This conceptualization is starkly different than that of the pro-CRT memes, and both are far from the “real” legal studies definition, as outlined above. As such, there is obvious political struggle in the fight to win the hegemonic, accepted definition of critical race theory — a definition which has little to do with its origins in critical theory and law. In this discussion on the political struggle over defining CRT, one particularly interesting meme to highlight is that of Cluster 27 (Figure 6). Metadiscourse on the Stakes of Defining CRT In this meme, a screenshot of a quoted tweet, two people are explicitly naming this struggle and pointing to its risks. By indicating that The Heritage Foundation, a conservative think tank known for their right-wing ideologies, is a top hit in searching for information on critical race theory, the original tweet author is pointing out that knowledge acquisition via the internet is deeply politicized, and without careful, critical consumption habits, people can be easily misled by seemingly authoritative information. The quote tweet, agreeing with this view, further interpolates into an ongoing debate in U.S. culture on whose “job” it is to educate on topics surrounding race and racism: people of color, who are exhausted by confronting this responsibility every day, but are also the people who have actual experience with racism; or white people, who are the ones who should be expending energy into acquiring knowledge to better educate themselves, without needing to further exploit the time and energy of people of color to do so (see, for example, Zheng (2021)). Both Twitter users, “naima, an improvisation,” and “Ramy” land in the same place: passing this responsibility to educate over to the internet is dangerous territory in a politicized information environment. If “some reactionary think tank” such as The Heritage Foundation is where people are gaining their knowledge because “Google is free” and no one else is providing this information, the struggle over defining these words–words which have actual policy impact, as seen through Youngkin’s Executive Order 1–is of utmost importance, and it appears that memes are one avenue in which this epistemological struggle occurs. Defining “Antiracism” These pro- and anti-CRT battles over definition, and the recognition in the metadiscourse that this is, indeed, a battle, is almost exclusively fought on the same grounds: that of antiracism. While there are some other nods to bigotry in its various forms, for example, transphobia as displayed in Figure 5, race and racism are unsurprisingly the main sites upon which definitions of critical race theory and its risks and/or benefits, are built. However, in a similar fashion to how “critical race theory” was defined to meet community needs rather than to reflect a “real” definition of the term, “antiracism,” too, is made into a fungible ideal constructed to meet the dire need of both sides of this argument to appear not-racist: a near necessity in 2023 America. According to Ferguson (2022), antiracism has suffered from a lack of coherent and accessible academic definition. Thus, she proposes a paraphrase of Black author and activist Ijeoma Oluo’s tweeted definition: “the commitment to eradicate racism in all its forms,” with a noted special interest in recognizing the difference between systemic and interpersonal racism. However, the tension we encounter in the analysis of these images is that, under this definition, both overt racism and a quieter “not-racism,” a term we will more clearly define below, is easily able to masquerade as antiracism to the undiscerning eye. In this way, the memes’ constructed definitions of antiracism become a technology through which racism is perpetuated. Bonilla-Silva (1997)’s oft-cited definition of racism demands recognition of both structural and interpersonal forms of racism, and it also demands acknowledgment of the difference between the two. Because these pro-CRT memes often fail to address structural racism, they fall short of forwarding truly antiracist ideologies even as they consistently present themselves as performing antiracism. These failures come through in what we are deeming “not-racism,” or essentially a focus on the interpersonal aspects of racism only, rather than on the structural ones. For instance, in Cluster 4 (Figure 2), the deflective “they” indicates several things all at once: first, that “we,” those who identify with this meme, are not like “them,” the racist Republicans; and second, that the stakes of this argument on critical race theory reside at the individual, interpersonal level. Each of these two implications constructs racism as something that happens within the hearts and minds of individuals, rather than at the structural level, and further, it absolves those who resonate with the meme from racism: “I am not a racist, because I support critical race theory.” This is not-racism in practice: a positionality that is not reflective of true antiracism, which is an ideology and practice that requires recognition of and action toward dismantling systems of racism, interpersonal racism, and the implicit bias that each of us holds. This is not to say that pro-CRT memes always failed at performing antiracism, either. There certainly were instances of successful acknowledgement of structural racism, such as in Cluster 7, which reads “If people attacked White Supremacy like they are attacking critical race theory, there would be no need for critical race theory.” However, the failures were all failing in the same way: by framing themselves and their ideologies as not-racist, and falsely equating that with doing antiracist work. On the other hand, anti-CRT memes wholly fail at performing true anti-racism, and they fail in many different ways: through rhetorics of racial neoliberalism, colorblind racism, and not-racism. Importantly, each of these tools of racism are constructed as antiracism, and sold to audiences as such. While not-racism presents a bit differently on this side of the aisle, the main takeaway is meant to be rhetorically the same: “we” are not the racists, “they” are. For instance, Figure 3 reads, in part, “What does critical race theory teach?...The only remedy to past discrimination is present discrimination.” This swiftly both dismisses racism as in the “past,” thus ignoring its structural persistence and constructs those that support critical race theory as the racists. In this way, both pro- and anti-CRT memes variably fail at performing antiracism, often forwarding what Blake et al. (2019) call “antiracist appropriation,” or a strategy that is “primarily concerned with deciphering who is a racist and who is not, rather than working to dismantle racism’s socially shared institutional and affective structures” (p. 23). By forwarding this claim, we do not mean to engage in an uncritical false balance (Rietdijk and Archer 2021) analysis here, as there is clearly one group that is getting closer to actual antiracism than the other: pro-CRT memes. It is important to note, however, that even pro-CRT memes are not actually accomplishing an antiracist agenda, as they are rather uncritical of the structural aspects of racism and choose to focus, instead, on interpersonal-level issues. Though using appeals to authority is not a groundbreaking rhetorical strategy and is, in fact, one of the pillars of Aristotelian rhetorical philosophy, the ways in which this ethos appears within these memes present an interesting finding: equally often, both pro-and anti-CRT memes deployed the imagery and/or quotes of Black people. Through circulating these images widely, those captured in these memes essentially stand in as Black spokespeople for each side of the argument, lending credence to the meme’s ideology–no matter the side of the argument–through the color of their skin. Appeals to Black Authority Anti-CRT memes that used this rhetorical strategy–all of which, notably, were produced and originally disseminated by the conservative Media Research Center (MRC)†–constructed these Black spokespeople as both authoritative in their experience and authoritative in their Blackness. Alveda King, Civil Rights Leader; Dr. Ben Carson, M.D. and former Secretary of Housing and Urban Development; and Dr. Carol Swain, Ph.D. and professor of political science and law appeared in these memes, each in visage and in quote (see Figure 7 for an example of these memes, each of which followed this aesthetic template). The embodied Black professional positionalities which these people inhabit make it difficult for pro-CRT advocates to argue against their claims—claims which invariably speak to the sure pitfalls of socioculturally adopting critical race theory—and thus a comfortable space of disseminating racism through the rhetorics of not-racism opens up. Using Black spokespeople to deliver implicitly or explicitly racist information has been a tactic used for decades to make news reporting (Entman and Rojecki 2007), campaign strategy, and political policy (Mendelberg 2001) appear notracist. The implicit suggestion is that if a Black person indicates something is not racist, it must not be. This, then, “bolsters [whites’] denials that racism still impedes the lives of African Americans” (Entman and Rojecki 2007, p. 106) and invites a level of assurance that they, too, are notracist. In the context of the memes analyzed for this study, these Black spokespeople are consistently reflecting a well-established space of Black conservative thought which taps into individualism, self-help, and egalitarianism as answers to discussions on racism (Lewis 2005). Critical race theory–the “real,” institutional one–actually warns against this very scenario: CRT argues that constructing Black spokespeople as people who can speak for the entire race is both essentialist and ignorant to the importance of intersectionality (Delgado and Stefancic 2023). While it is true that Black Democrats far outweigh Black Republicans in the electorate (Cox 2022), and thus pro-CRT memes that deploy Black spokespeople likely reflect a larger share of Black thought, it is still unproductive to count any single person as representative of a race of people. Despite this, however, constructing Black spokespeople through memes–Black spokespeople who are made to appear as the reasonable “Black voice”–was a way of building authority and “assurances” for those against CRT that they were not thinking in a racist way. While their Blackness operates, rhetorically, as an appeal to authority in a conversation about race, the idea of the “Black spokesperson” takes on a different function and meaning in these pro-CRT memes. Whereas anti-CRT memes were curating quotes from a very specific set of Black spokespeople, seeking out those who have said something condemning CRT, overlaying these quotes on an image of the Black orator, and circulating that image, those on the pro-side of the issue are more often amplifying already-existing media that Black people created. For example, the image of artist Jonathan Harris (Figure 1) was an organic, pre-existing photograph of the artist posing with his work–not a computer-generated, curated message that was created without his knowledge or consent. Similarly, the meme featuring Michael Eric Dyson’s thoughts (Figure 4) is simply a screenshot of a tweet he chose to write and publish on the internet–again, not something that an outside entity needed to create. The only exception is the meme featuring John Legend (Figure 9). This meme has a similar aesthetic to the anti-CRT memes in that it is a computer-generated image of Legend alongside a quote about Black parents needing to get involved in conversations around the banning of critical race theory. This meme represents just one way that anti- and pro-CRT memes perform “not-racism” in rhetorically similar ways, even if they have different end goals. [†] The Media Research Center (MRC) is more than simply a Facebook page, which is where many of these other memes are sourced. To the contrary, MRC is an entire conservative media network that self-describes its mission as being in accordance with "America’s founding principles and Judeo-Christian values." For more, see mrctv.org

Online Knowledge Production in Polarized Political Memes: The Tools of CRT Meme Production

Jews have become the perfect scapegoat for extremists on both sides A few months ago, before it was rebranded to ‘X’ by the richest moron on the planet, I left a comment on a Twitter post. The poster was an Asian woman, who had experienced some racism while living in Sweden. I’d had some negative encounters there myself, but because they were less about race and more about xenophobia, I made it clear that — aside from being Jewish — I was just a “white dude.” Shortly thereafter, someone replied: “No, you’re not. Jews aren’t white.” Another commenter agreed, and yet another called me a “parasite.” Interestingly, none of these comments violated Twitter’s new TOS. My main takeaway — aside from how terrible the site has become — was how some people’s lives are so shitty that the only thing they can cling to is their “whiteness.” As if it’s a flex. Newsflash: it’s not. But I digress. We’re not here to talk about sad white men. We’re here to talk about anti-semitism, which, unfortunately, has a much broader coalition behind it. While hating Jews has a long and storied history, it’s the people behind the hate who have changed However, from the looks of things, it seems as if it was always there, right beneath the surface. Just like how Trump allowed bigots and racists to show their true colors, October 7th did the same thing for antisemites. That’s why you see all the people tearing down posters of kidnap victims. Chants of “From the river to the sea” at rallies. Jewish college students having to hide from their classmates. There’s even a that just came out that said half of Gen Z sides with Hamas. Not the Palestinians, but Hamas itself. What does that tell you? poll I can hear the argument now: “But but but…Israel!” So let’s talk about that. No one is condoning what Israel and especially what Netanyahu have done. But how are Jews in America and other countries responsible? Why are they being , , ? harassed attacked killed Furthermore, the Israeli people are not the Israeli government. “But they voted for him!” Did you vote for Trump? Or Bush? Or Reagan? Was Putin elected, or Xi? Sometimes, we end up with leaders who don’t represent our best interests. Sometimes we’re dealt a crap hand. It doesn’t automatically make every single person in the country culpable. But that’s not even the real issue here. The real issue is that the situation is so complicated that it’s above most people’s pay grades Actual scholars and historians, people who have been studying this their entire lives, have trouble grasping all the details and nuances. What makes you think, in the weeks or months you’ve been following this, that you’re an expert? Furthermore, where did you get your information from? TikTok? YouTube? Again, it seems similar to the people on the other side, who learned everything they know from Facebook. Or Newsmax. Or Fox News. very In other words: You don’t know shit. You know what you’ve seen from 15-second video clips that are part of an algorithm specifically designed to trigger you, in order to keep you on the platform. That’s why both the far left and the far right are so misguided. And, frankly, why they’re so similar. Because in the internet age, nuance doesn’t exist. And it certainly doesn’t go viral. What does go viral? Extremes. Anything that makes your blood boil. Anything that further amplifies the feedback loop you’re stuck in. Anything that posits the simplest solution to a complex situation and completely eliminates any nuance or shades of grey. If you can throw in a convenient bogeyman for good measure, so much the better. For the right, it’s usually “wokeness” or other dumb buzzwords like CRT and “defund the police.” For the left, it’s pretty much anything Trump or one of his surrogates does or says. In most cases, there is a clear firewall between these messages. But once in a while, they align. In this instance, the Jews are the perfect punching bag for both the social justice warriors on the left and the racist goons on the right. The only good news — if you can even call it that — is that most people are not like this. While it might look like half of Americans are MAGA and the other half are “progressives,” both, in fact, are minorities. Most people (like myself and you probably) fall somewhere in the . We just don’t get as many clicks. middle So, where do we go from here? How do we fix what’s happening in Israel and Gaza? What’s the correct way forward? I have no fucking clue. And neither do you. And that’s fine. It’s not our job to figure all that out. Our job is to filter out all the clickbait and noise and to get the most accurate information possible. Our job is to elect leaders who will steer us through situations like this one with courage and compassion. Our job is to try and ensure the peace and prosperity of all individuals, regardless of their race, ethnicity, or religion. Lastly, it’s to stop blaming the Jews for everything. Between trying to pass for white, counting all our money, and controlling Hollywood and the media, we have enough on our plate already.

One Thing Both the Far Right and the Far Left Agree On? How Much They Hate Jews

143 Stories To Learn About Nft Marketplace

101 Stories To Learn About Nft Top Story

209 Stories To Learn About Mental Health

The International Subversives.

Computer vision has captured the interest of entrepreneurs all over the world, and for good reason: the capabilities of modern AI tech makes previously impossible ideas into real products. Detecting and classifying objects in photos and videos has found applications in many different areas and hundreds of systems — from security cameras with built-in facial recognition and detection of diseases based on x-ray scans to simple mobile apps. But enough with the intro, in this article I wanted to share the ‘behind the scenes’ of how a computer vision product is developed, especially a complex one. I have had an opportunity to work on Birdsy — a bird recognition app for people who want to see who visits their backyards while they are not looking. Birdsy is a complex, AI-powered app with real-time object detection and classification that needs to be able to run on weak hardware, detect bird types and sex with high accuracy. Given all that, the road from the initial idea to publishing the app in app stores was complicated — and captivating at the same time. We have faced many hurdles — both from the business and development side of view — that I’ve decided to share in one place to hopefully help entrepreneurs and AI developers who might be facing the same project. The task at hand Birds have had millions of years of evolution to blend in perfectly with their environment to avoid predators and, in this case, bird watchers, making it harder for them to admire wildlife. While observing certain bird species face to… beak can be problematic, monitoring them through a video camera from the comfort of your home is a lovely way to enjoy our winged friends, especially if AI removes the need for looking through hours of video footage and sends you alerts when a bird has entered the camera’s view, automatically detecting what bird species it is. There are two parts to Birdsy: Object detection — a neural network in charge of detecting the presence of a bird or a mammal Object classification — a neural network in charge of determining which bird or mammal species was captured Low-quality cameras for object detection To make the service more approachable and easy to use, any camera can be used for bird watching. This is where we ran into the first problem: low-quality cameras as they are the most affordable and the most widespread. While the ‘no camera limit’ is great for users, it presented a challenge for us, as the object detection model runs using the camera chipset. Where someone gets a good deal, others get the short end of the stick, in this case ‘others’ being our CV developers. Working with a cheap camera means working with a cheap chipset which makes it impossible to use the default neural network architecture. Compared to a top-of-the-line, gold standard of computer vision video cameras (NVIDIA Jetson Nano) which allows using around 120 layers of default YOLO v4, the cameras we had to work with allowed only for 22 layers. Where a full YOLO v4 neural network provides great recognition results, a stripped-down version performs poorly. We have tested both and were unpleasantly surprised with how low the model depth was when running it using a cheap chipset. Testing default YOLO v4 vs Reduced We started with training the default YOLO v4 model and testing it on the customer’s dataset The results we achieved were satisfactory - 95% mAp, which in the world of computer vision is more than enough to launch a model into production. After retraining the model to fit the camera’s parameters, the detection quality dropped significantly. But where machines fail humans advance. We have tested the neural network on test data and visually evaluated false positives and false negatives. This highlighted where the network lacked knowledge and where it made the most mistakes. We ain't nothin but mammals, or false positives The network was eager to detect people, especially people’s hands, as animals (and we don’t blame it, humans ARE animals after all). While, from a biological point of view, this is correct, the end-user is more interested in looking at birds than their neighbors, so we had to work on teaching the network to ignore people and focus on birds and mammals instead. To do this, we have added negative examples, including pictures of people at various angles as well as human hands. What lurks in the dark The cameras have two modes: regular daytime mode which produces full-color images, and nighttime infrared mode, which produces black and white images. When the camera switched to infrared, the model produced a lot of false positives: tree leaves moving in the wind insects, like moths, that are attracted by a warm camera fountains Users would be less than happy to be woken up by a notification, get excited to look at an owl or a fox, but end up looking at a recording of a moth banging its body against a camera lens. In order to reduce sleep interruptions only to a minimum, we collected instances of false positives in nighttime settings and marked them by hand. Instagram vs Reality Ever heard of social media being called a ‘highlight reel’ where people present the best version of themselves? Who knew the same could be true for wild animals. Photos of birds that can be obtained from open sources, like Google Images and Youtube videos, are usually high quality, very sharp, and depict specimens at their best, looking at the camera or at least positioned front on, with nothing in between the bird and the camera obstructing the view. Reality is not always as pretty. Cameras produce low-quality images that can make it hard to understand what’s going on even for a human eye, bad weather conditions like rain, snow, or dust can obstruct the view, and we are sure birds sense when someone wants to capture them and position themselves in the most ridiculous way possible. The dataset that the client provided (consisting of shart images found on the Internet) was not of very much use for this project. We needed to collect a set of images of birds in real conditions using the client’s cameras to show the model what birds look like, and not how they are presented on social media. really What we have now So, after doing all of the above: manually going through every instance where the network was wrong teaching the network to ignore people and not detect them as birds or mammals working with black and white images and false positives collecting a dataset of birds in real-life conditions, under various angles, and in different weather conditions We have managed to achieve a 97,5% mAP for object detection. This is a very high result for a computer vision model as the unwritten rule for any CV model going into production is to have over 94% mAP. Achieving perfection (almost) While the results we have achieved now are more than enough to be used in the final product, there’s still room for improvement. the network mistakes flowers for birds birds’ view from the back is still sometimes a mystery birds partially covered by random objects birds close to the camera After enough images for each group are collected, we expect the mAP to increase and reach 98,5%. Image classification model The next step in getting to know your backyard visitors is to pass the image with a bird to an object classification model. Its goal is to recognize the bird species and its sex. As some bird species live exclusively on certain continents, we have decided to create two models - one for those who live in North America, and one for those in Europe. One step at a time: a lesson in multistage neural networks Initially, the problem of object classification was solved using a ‘head-on’ approach: the network was shown photos of all the different species, both males and females, from which it tried to learn what they look like and how they differ from each other. This resulted in very poor accuracy scores, in other words, the network made a whole lot of mistakes when identifying bird and mammal species. The network was trying to learn too many aspects at the same time. Many bird species look very similar to each other and differ from one another by a single patch of differently colored feathers or a differently shaped beak. Retaining all of this information, along with what different sexes of the same species look like, is too difficult under given circumstances. The network would often mix up bird species while determining the more broad bird type correctly. For example, the difference between a hooded warbler and a kentucky warbler in a patch of black feathers: The network would label a hooded warbler as a kentucky warbler, producing a wrong result, but being generally correct: both ARE warblers. For the sake of time, the client decided it was more important to detect the overall bird type rather than its particular species, so that’s where we started. After model evaluation, we have decided to implement a multistage approach: all bird species were grouped into more general categories: eagles, sparrows, crows, pigeons, etc. the network would first determine the overall bird type the second step is to detect which species it is the last would be to determine if the bird is male or female By grouping the bird species, we have managed to decrease the number of classes from 98 to 49, which greatly improved the accuracy score as the network simply didn’t have as many classes to choose from. Practice makes perfect, even for machines When you come across a new concept, you read books or watch educational videos to grasp it. If you fail, you ask your friend to explain it to you or you visit a seminar on the topic. In other words, you try to accumulate more information about it to understand it better. The same goes for neural networks. The better you need one to understand what a warbler looks like, the more images of warblers you need to show it. The more data it has looked at, the better accuracy scores will be. The multistage approach we’ve chosen not only improved the accuracy of the object classification model but also made it possible to analyze the dataset and determine where the network lacked learning data. The ugly duckling problem After the object classification model was launched, we were surprised to receive results that were much worse than what the tests showed us: the model could not determine the bird species or type correctly. The problem ran deeper than this: our computer vision developer in charge of the whole project, who has learned all of the bird species himself while working on it, also failed to determine what the birds were when he received the images incorrectly labeled by the network. It turned out that July is not the best time to launch a bird classification model as it is the time where teenage birds learn to fly and leave their nests. Remember the ugly duckling story? Well, it’s true for most birds, fledglings look nothing like adult birds, and it is hard to know what bird you are looking at if it’s still young. We have collected images of young birds over the summer and plan on training the classification network to determine the different bird species at different ages. Teamwork makes the network work Birdwatchers are a passionate bunch, they know how to identify a bird by the shape of its single feather. They possess knowledge that our classification network dreams of having, so why not bring the two together and form a bird-loving alliance the world has never seen before? Currently, the classification network doesn’t just tell the user the bird species, it shows the degree of confidence along with other guesses. The user can confirm the network’s guess or correct it, thus helping us train it - one bird at a time. After running the user feedback system for 3 months, we have collected over 20 thousand images. This data is invaluable to us since the photos are made in real-life conditions (bad weather conditions, at night, etc.) and are marked up by experts. When you stare into the abyss, the abyss chirps back at you It is worth noting that during this project, we have become the bird experts ourselves. Looking at birds all day long, while essentially educating a virtual child on all the little differences between different types of sparrows, makes one an instant platinum member of the birdwatching community. If all else fails, our CV team members can easily find themselves in ornithology. On a serious note, looking through thousands of images of birds, be it for dataset markup or analyzing where the network makes the most mistakes, we have delved deep into this project and came out the other end not only with a bunch of bird knowledge but a better understanding of how complex image recognition and classification systems work, how to best implement them, how to analyze a large dataset and find its weak points. This project was invaluable for us as an opportunity to research and work with the latest computer vision technologies, work with real-time customer feedback, and polish our problem-solving skills when working with outdated code.

Creating a Bird Detection AI: From Ideation to Product Launch

একটি পাখি সনাক্তকরণ এআই তৈরি করা: আইডিয়া থেকে পণ্য লঞ্চ পর্যন্ত

Erstellen einer Vogelerkennungs-KI: Von der Idee bis zur Produkteinführung

Creación de una IA para la detección de aves: desde la idea hasta el lanzamiento del producto

Créer une IA de détection d'oiseaux : de l'idéation au lancement de produit

बर्ड डिटेक्शन एआई बनाना: विचार से उत्पाद लॉन्च तक

鳥検出 AI の作成: アイデアから製品発売まで

조류 감지 AI 만들기: 아이디어 구상부터 제품 출시까지

Criando uma IA de detecção de pássaros: da idealização ao lançamento do produto

Создание ИИ для обнаружения птиц: от идеи до запуска продукта

Kuş Algılama Yapay Zekası Oluşturma: Fikir Geliştirmeden Ürün Lansmanına

Tạo AI phát hiện chim: Từ ý tưởng đến ra mắt sản phẩm

创建鸟类检测人工智能：从构思到产品发布

WOKEN

iamwoken.com

-- employees

Since n.d.

COMPANY RANKING

WOKEN

EVERGREEN INDEX #11779

WOKEN WIKI

Apply to get your Tech Company News Page live on HackerNoon.com today!

WOKEN

iamwoken.com

-- employees

Since n.d.

COMPANY RANKING

WOKEN

EVERGREEN INDEX #11779

RELATED COMPANIESRANK

STORIES ABOUT WOKEN

WOKEN WIKI

Apply to get your Tech Company News Page live on HackerNoon.com today!