We obtain more highlight information through underexposure and more shadow information through overexposure. By combining images with different exposures, we obtain an HDR image with more details.
brandonlouie
I'm interested to know more about the solid blue bars on the last row of images. What's interesting to me is that we can see the blue borders swap between appearing at the top of and at the bottom of the image. To me, it makes more sense that as the blue border grows (as evident when viewing the images from left to right) that it grows in place, but it's interesting to me why they might flip. Does anyone know why this is the case?
buggy213
i actually worked with this dataset for the 180 final project. as it says in the caption, the black borders are an artifact of aligning all of the images. i think this didn't really cause problems with the hdr exposure calculation, as you can choose sampled locations which are not inside of these problematic areas
olliep24
I assume that this method can only work for still images because there are multiple photos that need to be taken of the same scene. I wonder if there is a way to capture all the range faster. A possible solution could be using multiple cameras, but the perspectives of the cameras would be different for the scene, which would need to be adjusted for.
brianqch
I think if you use a decently quick shutter speed, you might still be able to capture HDR images for objects that are moving. There might be some artifacts when blending the images together to capture more details but I know that a lot of cameras out there have a continuous shooting mode. That might be able to help?
sebzhao
I heard from photography class that the reason you can not unexpose overexposed areas and expose underexposed areas when they reach true black or true white is because the information is lost, and it's cool to see what is actually happening is that the pixels saturate. Also super cool that these true blacks and true whites actually are helpful by using multiple exposures.
We obtain more highlight information through underexposure and more shadow information through overexposure. By combining images with different exposures, we obtain an HDR image with more details.
I'm interested to know more about the solid blue bars on the last row of images. What's interesting to me is that we can see the blue borders swap between appearing at the top of and at the bottom of the image. To me, it makes more sense that as the blue border grows (as evident when viewing the images from left to right) that it grows in place, but it's interesting to me why they might flip. Does anyone know why this is the case?
i actually worked with this dataset for the 180 final project. as it says in the caption, the black borders are an artifact of aligning all of the images. i think this didn't really cause problems with the hdr exposure calculation, as you can choose sampled locations which are not inside of these problematic areas
I assume that this method can only work for still images because there are multiple photos that need to be taken of the same scene. I wonder if there is a way to capture all the range faster. A possible solution could be using multiple cameras, but the perspectives of the cameras would be different for the scene, which would need to be adjusted for.
I think if you use a decently quick shutter speed, you might still be able to capture HDR images for objects that are moving. There might be some artifacts when blending the images together to capture more details but I know that a lot of cameras out there have a continuous shooting mode. That might be able to help?
I heard from photography class that the reason you can not unexpose overexposed areas and expose underexposed areas when they reach true black or true white is because the information is lost, and it's cool to see what is actually happening is that the pixels saturate. Also super cool that these true blacks and true whites actually are helpful by using multiple exposures.