3D Scanning with Kinect v2

Uncategorized

3D scan of full body

3D scan of full body

By on Nov 9, 2015 in Uncategorized | 0 comments

For our next adventure, we will be working on merging multiple 3d scans into a full body model. When we are done we will upload a video like the others. Alignment Tips Keep in mind that if you keep aligning scans wrong then by the time you finish you may have extra features or missing features. After a quick rush through, the only flaw in the scan was double elbows. Extra features mean that 2 scans weren’t properly aligned before creating the final mesh, same thing happens for missing features or thin arms. This did happen with the rush alignment, I just wanted to make sure there was enough model data to complete the final model, I was surprised that it turned out so well on the first try while being sloppy. So, pick points that are easily distinguishable and effect the final outcome of the render- it doesn’t have to be 100% accurate, just close enough so the fine-tune alignment can work. In the above image is circled in red areas that are easily distinguishable and effect the final mesh. Who knew those picture matching games would actually help out with real world work. Because the Kinect 2 distortion, more scans actually makes it more difficult to get an accurate alignment, the fewer scans the easier the alignment process will go – it’s a balance between capturing enough data to get the 3d model, but not too much data that it’s hard to align properly. Another Picture! Well, this next video will probably be a long one, getting a full model stitched together can be tricky. I’ve had to discard several captures as too many captures was making things...

Read More
3D scan of shoulders and head

3D scan of shoulders and head

By on Nov 5, 2015 in Uncategorized | 0 comments

While getting the scans merged together for a head/shoulder scan, we’ve found a couple problems that will need to be worked around. Hair scans poorly While aligning two scans we noticed that the hair was not lining up, and that it actually couldn’t line up correctly because it became deformed -depending the angle scanned. The solution is simple, certain angles will give lower quality scan data, the scan with less surface area and would contribute less to the model seems to be the bad data. Trimming the false hair would improve the quality of scan without losing any data (the complete area of hair shows up on the good scan). Fewer features to map the back with While trying to attach the back together, it’s considerably more difficult to attach the back together. The front has very distinguishable marks between scans (nose, chin, eyebrows) – where the back has shirt wrinkles, general spots on shoulders. We will be digging into this more to try to find the best approach to aligning the back. It looks like it will involve first aligning the front of the scan, and then attaching the back to it. Initial Results We were able to overcome the initial obstacles and will be working on the video today....

Read More
3D Scanning a Face

3D Scanning a Face

By on Nov 4, 2015 in Uncategorized | 0 comments

Goals What I consider a good facial scan is: You can tell who it is – enough features to distinguish between different people. The nose looks right. The shape of the head is correct. There are no extra features added like double cheek bones. This is just for the face, between ears, chin and forehead. Initial findings The Kinect V2 cannot scan closer than 50cm, so trying to get the face the cover the majority of the scan area is impossible; this also works out well because the edges of the scan area are not good to work with anyways. The side facing towards the scanner is good, but the side of the face that is angled away from the scanner will be lower quality data, I think I will have to trim it out to get a good scan/alignment. When scanning, quality is related to holding still and the distance of the scan. The closer you are, the more detail there will be- but the further away, the less movement will interfere with the scan. How to get good quality alignment on faces Get 3 good Scans: Front, left and right side at (45 degree angle). Pick easy alignment points; Between the eyebrows, tip of the nose, under the bottom lip, check bones. Conduct the alignment process and convert into a printable mesh: Here is a video with the process...

Read More
Working on Alignment: Walkthrough

Working on Alignment: Walkthrough

By on Nov 3, 2015 in Uncategorized | 0 comments

To get a 3D scan of something, you take pictures all around it. Those pictures will naturally overlap. Our program helps you align these 3D pictures. Considering that properly aligning 3D snapshots is a pivot point between a good or bad 3D model, we are working this week on what the key points are in creating a good alignment the first time and what to look out for. (We will not be focusing on color as we get these alignments done; rather, we will focus on the specific techniques for getting a colorless 3D model produced at the best match to real life.) Alignment points When aligning two scans, you need to choose distinguishing marks on the scans to match. For example, we could match the left ripple in the curtain from the first scan to the left ripple in the curtain on the second scan. The program merges those two points together. However, to make this work, you need: 1) distinguishable points, 2) enough points. Distinguishable points Two of our office scans were of flat, empty walls, so there were no distinguishable points to match.  This makes alignment immensely difficult. The three scans we were able to merge effortlessly on the first try (see the picture below) were easy to do because there were multiple points we could easily distinguish to align the scans with (the fan blades, the curtains). Enough points In the picture above, we had distinguishable points, like the curtains, the fan blades, and the door. However, if we marked only one of those distinguishing features, the two scans would not align very well. To align well, we mark multiple sets of points. So, in scan A, I could mark the left ripple in the curtain, the far right fan blade, and a point on the door. On scan B (which overlaps some of scan A), I would mark the same three points so the picture would align. Having enough points spread across the scan aligns the snapshots correctly. Kinect distortion around edges We’ve noticed that the Kinect seems to curve more near the edges of the scan. This is most likely a calibration issue with the Kinect (the factory calibration does not seem to be 100% accurate), but it could also be an...

Read More
Note: Scan from Life is in no way affiliated with Microsoft. We are an entirely separate company that has created a product that is dependent on a Microsoft owned product.