1. Depth-based Feature Recognition

    00:15

    from Seth Gibson / Added

    139 Plays / / 0 Comments

    This little bit was cool enough that I didn't want it to die without being forgotten, here we're using recolored depth data to generate an LBP cascade and using OpenCV to track the features. All Code and Content by Seth Gibson

    + More details
    • Face Detection try #1 (130616)

      00:22

      from tuwuhs / Added

      50 Plays / / 0 Comments

      Face detection using Haar cascade classifier. Scale parameter 3.0 (detection is performed on image 1/3.0 of the original size). Built using OpenCV 2.4.5 and openFrameworks 0.7.4, on Linux x86-64 platform.

      + More details
      • CV Dazzle: ITP Thesis Demo Look #1 (after)

        03:27

        from Adam Harvey / Added

        10.5K Plays / / 4 Comments

        Jen, after CV Dazzle For before visit: http://vimeo.com/34545827 Model: Jen J. Hair: Pia Vivas Makeup: Leigh B. More info: http://cvdazzle.com

        + More details
        • OpenCV Visualization: Eye Cascade

          01:29

          from Adam Harvey / Added

          304 Plays / / 0 Comments

          Visualization of the detector shapes used in the "eye.xml" cascade file for OpenCV. (Compared to the face cascades, the eyeglasses cascade is way more visually interesting) Each cascade can have 20 or so stages. The beginning stages are easier to pass, as noted by the larger shapes. As the algorithm progresses through each stage, some image subregions are quickly rejected. This allows the algorithm to quickly scan the image and not waste on unimportant areas. While towards the end of the visualization, the shapes will become smaller, more complicated and more picky. The Viola Jones algorithm compares the difference between the dark and light areas defined by the shapes and measures whether this difference falls within a threshold. Rendered at 12fps Part of my research for cvdazzle.com

          + More details
          • OpenCV Visualization: Eye Glasses Cascade

            01:11

            from Adam Harvey / Added

            1,187 Plays / / 0 Comments

            Visualization of the detector shapes used in the "eyeglasses.xml" cascade file for OpenCV. (Compared to the face cascades, the eyeglasses cascade is way more visually interesting) Each cascade can have 20 or so stages. The beginning stages are easier to pass, as noted by the larger shapes. As the algorithm progresses through each stage, some image subregions are quickly rejected. This allows the algorithm to quickly scan the image and not waste on unimportant areas. While towards the end of the visualization, the shapes will become smaller, more complicated and more picky. The Viola Jones algorithm compares the difference between the dark and light areas defined by the shapes and measures whether this difference falls within a threshold. Rendered at 12fps Part of my research for http://cvdazzle.com

            + More details
            • OpenCV Visualization: Frontalface Alt Tree Cascade

              11:46

              from Adam Harvey / Added

              451 Plays / / 0 Comments

              Visualization of the detector shapes used in the "frontalface_alt_tree.xml" cascade file for OpenCV. Each cascade can have 20 or so stages. The beginning stages are easier to pass, as noted by the larger shapes. As the algorithm progresses through each stage, some image subregions are quickly rejected. This allows the algorithm to quickly scan the image and not waste on unimportant areas. While towards the end of the visualization, the shapes will become smaller, more complicated and more picky. The Viola Jones algorithm compares the difference between the dark and light areas defined by the shapes and measures whether this difference falls within a threshold. Rendered at 12fps Part of my research for cvdazzle.com

              + More details
              • OpenCV Visualization: Frontalface Alt2 Cascade

                01:27

                from Adam Harvey / Added

                235 Plays / / 0 Comments

                Visualization of the detector shapes used in the "frontalface_alt2" cascade file for OpenCV. Each cascade can have 20 or so stages. The beginning stages are easier to pass, as noted by the larger shapes. As the algorithm progresses through each stage, some image subregions are quickly rejected. This allows the algorithm to quickly scan the image and not waste on unimportant areas. While towards the end of the visualization, the shapes will become smaller, more complicated and more picky. The Viola Jones algorithm compares the difference between the dark and light areas defined by the shapes and measures whether this difference falls within a threshold. Rendered at 12fps Part of my research for cvdazzle.com

                + More details
                • OpenCV Visualization: Frontalface Alt Cascade

                  02:58

                  from Adam Harvey / Added

                  241 Plays / / 0 Comments

                  Visualization of the detector shapes used in the "frontalface_alt.xml" cascade file for OpenCV. Each cascade can have 20 or so stages. The beginning stages are easier to pass, as noted by the larger shapes. As the algorithm progresses through each stage, some image subregions are quickly rejected. This allows the algorithm to quickly scan the image and not waste on unimportant areas. While towards the end of the visualization, the shapes will become smaller, more complicated and more picky. The Viola Jones algorithm compares the difference between the dark and light areas defined by the shapes and measures whether this difference falls within a threshold.

                  + More details
                  • OpenCV Visualization: Fronaltface Default Cascade

                    04:03

                    from Adam Harvey / Added

                    2,676 Plays / / 1 Comment

                    Visualization of the detector shapes used in the "frontalface_default.xml" cascade file for OpenCV. Each cascade can have 20 or so stages. The beginning stages are easier to pass, as noted by the larger shapes. As the algorithm progresses through each stage, some image subregions are quickly rejected. This allows the algorithm to quickly scan the image and not waste on unimportant areas. While towards the end of the visualization, the shapes will become smaller, more complicated and more picky. The Viola Jones algorithm compares the difference between the dark and light areas defined by the shapes and measures whether this difference falls within a threshold.

                    + More details
                    • Visual Servoing with OpenCV and HaarCascade

                      01:12

                      from David J Barnes / Added

                      147 Plays / / 0 Comments

                      The purpose of this video is not to illustrate the OpenCV haarcascade haarcascade_frontalface_alt2.xml. Although this cascade is used, I use it as a means to demonstrate visual servoing. Once a suitable MCvAvgComp object is found using the DetectHaarCascade method, the X position is obtained. A new rectangle is draw around the object and the X coordinate is transmitted to the MCU via USB serial. Upper and lower threshold parameters can be entered. It is between these thresholds that the servo constantly seeks as an optimal position. Currently only a horizontal servo is used.

                      + More details

                      What are Tags?

                      Tags

                      Tags are keywords that describe videos. For example, a video of your Hawaiian vacation might be tagged with "Hawaii," "beach," "surfing," and "sunburn."