Wednesday, March 10, 2010

Fooooo: Video Search Engine

Fooooo is a very popular and widely used video search engine in Japan. It is available in eleven (11) languages; the default is Japanese, and its users are able to search 220 million video results from some ninety-four (94) video providers at the time of my writing, showing itself to be an ambitious video search engine. Apparently, with the capability of searching 220 million video results on the Web, a user can search whatever video she want from this site, including some big-name sources like YouTube, Dailymotion, MySpace, NicoVideo, Nifty, Ameba, FC2, Oricon, YouKu, Tudou, etc. As stated on the front page of Fooooo, it is considered to be the biggest video search engine on the Web now.

On the Fooooo site, although the design of the English site much like the default Japanese site, but potential users can’t sign-up to the site and become a member as the membership privilege can only be found on the Japanese site. However, these both Japanese and English sites, users will not find any advertisement on the landing page, not like the other nine (9) language sites, a horizontal Google AdSense advertisement can be found on the bottom of the page.

One of the best parts of Fooooo is that the default video size you watch via Fooooo is slightly bigger than YouTube, providing a convenience that has proven quite attractive for the video avid users. Also, users have the option to adjust the video size from original to medium and to large size.

Fooooo is a very powerful video search engine, not only you can find what you want to watch on it, you also can narrow your results to certain category, or filter it by channel or duration. According to the official blog of Fooooo, they’ve statistics showing that Internet Explorer is used by 86.86 percent of its users while 8.77 percent using Firefox, and the most popular search term in this September is “arashi”, as shown in the picture below.

GazoPa

gazopa_image

GazoPa is an image search engine that launched recently on the Web. With most of the popular search engines such as Google, Baidu, Yahoo!, Microsoft’s Live Search have incorporated image search as one of the search services rendered, GazoPa is a visual image search engine that solely generating results for similar images.

GazoPa, currently in invitation-only beta phase, it enables users to perform similar image search for more than 50 million images on the Web through its proprietary technology. On the GazoPa site, you can search images by uploading your own image file, drawing, or type a keyword, and even if you do not have any inventory of images, you can type a URL link to GazoPa’s search bar. GazoPa will then generating all the similar images based on the characteristics of color and shade extracted from the image itself. The results generated might not in the form of images, it can also be appeared as videos, as GazoPa supports video thumbnails search by using image.

When testing the accuracy of GazoPa’s image search service, I’ve uploaded a satin ribbon photo. Not only it did not generate results based on the same category, it also did not render my desired results, i.e. a bunch of products (See the picture shown in below). Based on the search results, perhaps I would be understood GazoPa more if I knew that GazoPa’s image results are determined on the similarity of images instead of the sameness of images in the first place. On contrary, I performed an image search with the same item using the aforesaid TinEye, zero results appeared at the time of my writing. However, as I used GazoPa, it will definitely generate all the images available in its database based on the similar color, in this case the images with the light gray color.

In using GazoPa, users are allowed to search through playing tools such as trimming, region fill, or eye dropper tool that act as filters. There is also a search option in which you can click any one of the parameters, such as color, moderate1, moderate 2, shape, face, image size, monochrome only, or omit some images, thus you’ll get to learn the different search aspects rendered by GazoPa.

At this moment, GazoPa seem to be able to search images available in MySpace, YouTube, Twitter, Wikipedia, Yahoo!, Alibaba, but my general sense is that its actual search will need to be more robust to be most useful.

To know more about GazoPa, it is headquartered in Japan and being backed by Hitachi.

gazopa2_image