"Can I call you 'babe'?"

Recently I was chatting with some colleagues, and we were discussing a researcher joining our group. One person started to say, "And this new girl-,"  looks at me, "-Oh, or should I say, 'woman'?"

All the men in the room turned to look at me.

"Definitely 'woman'. I would only say 'girl' if I was speaking about a child."

This question actually sparked a good, productive conversation. Sometimes these situations can get awkward when one is asked to represent their entire sex, but I really appreciate it when people care enough about their language to ask.

Photo by Occhiovio

I also really appreciate when people correct me on my language usage. Lately I've been trying to not use ableist language in my speech, and am glad to be called on it. It's really difficult to deprogram yourself once you establish certain patterns, but it's worth it.

Why? Because language shapes thought. While the Sapir-Whorf Hypothesis is unlikely to be true in its strongest form (language determines thought), it is most certainly true in its weak form (language influences thought). Recently cultural neuroscience has been making a splash, showing things like how people raised in individualistic or collectivist cultures will respond to the same words in completely different ways. Other studies have suggested that education level affects how people differ in the way they consider concepts such as freedom and choice.  And who knows if any of the studies we've conducted over the past century have external validity outside WEIRD (Western, Educated, Industrialized, Rich, and Democratic) populations.


Words matter. So when talking to someone who represents a group that you are not a part of, it's not a bad idea to ask them what term they'd prefer you use when you refer to them or their group. Much better that then putting your foot in your mouth!

the little man....

is finally here.....and i can't get enough of him :)

we thought we had the perfect baby...the first night home from the hospital, he barely woke up to eat....the second night, a completely different story. we are a little sleep deprived (warren just made a grill cheese and forgot to put the top piece of bread on). here are some pictures of walker's room; promise to be more bright eyed later this week...well, maybe :) thanks so much for all of your well wishes for the delivery....it was very easy, thankfully.

*ps. taylor LOVES her baby brother, which is a good thing b/c he looks just like her!

KOPI Herbal - Kopi kaya dengan kandungan herbal

Kopi Herbal merupakan salah satu produk minuman untuk meningkatkan stamina dan kesehatan. Kopi ini mengandung berbagai macam herbal berkhasiat diantaranya:
1. Mesoyi, Kapulaga, serimpi, dan secang untuk mencegah masuk angin.
2. Jahe Merah dan Lengkuas merah untuk kebugaran tubuh.
3. Daun Dewa untuk melancarkan aliran darah.
4. Ginseng, dan Euricoma Radix (pasak Bumi) untuk stamina,
5. Pegagan untuk revitalisasi otak dan kulit (dimana kadarnya dibuat tepat sehingga efek sedatifnya berkurang)
6. dan Tribulus untuk menjaga organ reproduksi kita;
7. Tapak liman dan daun sendok untuk menjaga ginjal dan juga menjaga tekanan darah.

semuanya komposisi itu kita campur dengan kombinasi yang tepat dan kita extraksi sehingga bau-bau herbalnya menjadi berkurang tanpa mengurnagi khasiatnya.Disamping itu kopi yang kita gunakan adalah kopi arabica extract kita pilih agar kopinya harum, serta cream yang digunakan adalah non dairy creamer atau creamer nabati non kolesterol. Segera nikmati ....

Anda tertarik.. silakan contact : 021-96120932 (budi prakoso) untuk pemesanan. Harga 35rb/10sachet. MIn.order 4 kotak.

Object detection vs. Semantic segmentation

Recently I realized that object class detection and semantic segmentation are the two different ways to solve the recognition task. Although the approaches look very similar, methods vary significantly on the higher level (and sometimes on the lower level too). Let me first state the problem formulations.

Semantic segmentation (or pixel classification) associates one of the pre-defined class labels to each pixel. The input image is divided into the regions, which correspond to the objects of the scene or "stuff" (in terms of Heitz and Koller (2008)). In the simplest case pixels are classified w.r.t. their local features, such as colour and/or texture features (Shotton et al., 2006). Markov Random Fields could be used to incorporate inter-pixel relations.

Object detection addresses the problem of localization of objects of the certain classes. Minimum bounding rectangles (MBRs) of the objects are the ideal output. The simplest approach here is to use a sliding window of varying size and classify sub-images defined by the window. Usually, neighbouring windows have similar features, so each object is likely to be alarmed by several windows. Since multiple/wrong detections are not desirable, non-maximum suppression (NMS) is used. In PASCAL VOC contest an object is considered detected, if the true and found rectangles are intersected on at least half of their union area. In the Marr prize winning paper by Desai et al. (2009) more intelligent scheme for NMS and incorporation of context is proposed. In the recent paper by Alexe the objectness measure for a sliding window is presented.

In theory, the two problems are almost equivalent. Object detection reduces easily to semantic segmentation. If we have a segmentation output, we just need to retain object classes (or discard the "stuff" classes) and take MBRs of regions. The contrary is more difficult. Actually, all the stuff turns into the background class. All the found objects within the rectangles should be segmented, but it is a solvable issue since foreground extraction techniques like GrabCut could be applied. So, there are technical difficulties which could be overcome and the two problems could be considered equivalent, however, in practice the approaches are different.

There arise two questions:

1. Which task has more applications? I think we do not generally need to classify background into e.g. ground and sky (unless we are programming an autonomous robot), we are interested in finding objects more. Do we often need to obtain the exact object boundary?

2. Which task is sufficient for the "retrieval" stage of the intelligent vision system in the philosophical sense? I.e. which task is more suitable for solving the global problem of exhaustive scene analysis?

Thoughts?

Communication Styles and Skin Thickness

I've started this blog, mainly, because of some recent articles in the NY Times by John Tierney regarding the reasons why their are so few women in STEM. While some of his writing contained bits of truth, mostly it was a thinly veiled opinion piece that the editors should never have put in the 'science' section.

But I have no desire to attack Tierney or the Times editorial staff beyond voicing my displeasure at both. What I want to discuss here is why women really leave STEM.

It's all about the people, and their communication styles.

Photo by Articulate Matter

Throughout my life, I have participated in many activities where there are hardly any women - CS departments, engineering companies, tinkering groups, ham radio, some sports. Aside from occasional pleasant surprises in my professional life (such as once attending Grace Hopper), I'm often the only woman in the room. But generally I haven't found this to be a problem. I am lucky in this regard, I know this can be a problem for many women, especially at first. But over time I've learned strategies that help me to not feel intimidated giving a talk in front of a room full of men, to not feel worried about speaking up in meetings, and to hold my
own arguing to the death about software design.

But the one thing that has, on occasion, made me want to leave technology entirely and open a bakery are people who come across as jerks. People who are deaf to the tone of their affect, who do not understand that their mannerisms would be considered rude by most people, who act seemingly unconcerned about how others might feel in reaction to what they say. The good news is these people are usually gender-egalitarian in their thorniness, but I think sometimes for some women, enough encounters like this make you want to leave the rose garden.

"You need to grow a thicker skin" and "Don't take it personally" are comments I heard early in my career, and still hear as advice given to young women embarking on theirs. Women are told, particularly in academic science, that if they want to be successful they need to be able to handle the beatings that can come in a manner that is unquestionably brash, rude, and humiliating. We are told to not cry in front of others, we are told to not lose face, and ultimately, we are told to act like men. (Except not too much, because then you become unlikeable, and that's also a career killer for STEM women. Surprise!)

The problem is that most women I know have much lower emotional pain thresholds when it comes to their professional lives than most men I know. And while other professions have their share of people interaction problems, they seem to be less tolerated to the degree they are in STEM. I've attended many a talk where someone in the audience interrupts the speaker, repeatedly, to nitpick. Nobody will ever pull Dr. Jones aside and say, "You had some good points, Dr. Jones, but did you really have to be so rude making them?" The problem is, Dr. Jones is not going to notice the quiet sighs and subtle eye rolls every time zie acts up during seminar. Dr. Jones does not get hints. Dr. Jones may or may not respond to directness, but by the time others in the group have worked up the gumption to say something, it's years too late. The humiliated person is long gone, from the organization and perhaps from science itself.

This is a problem that needs to be addressed on mutiple levels. Yes, wronged people need to rise from the ashes, get their game on, and fight back. But everyone else needs to stop acting so tolerant of brash behavior in science. And for people who act brashly, they need to learn, as much as they are able, some more positive interaction behaviors.

I don't think most people are jerks. I just think many of them are completely unaware of how they are coming across.