# Optimization Parameters Tuning

**URL:** <https://shapeworks.discourse.group/t/optimization-parameters-tuning/393>\
**Category:** Uncategorized\
**Created:** [February 20, 2026, 5:08pm UTC](https://shapeworks.discourse.group/t/optimization-parameters-tuning/393 "2026-02-20T17:08:44Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![blah4](https://avatars.discourse-cdn.com/v4/letter/b/7ea924/32.png) [@blah4](https://shapeworks.discourse.group/u/blah4)\
**Post date:** [February 20, 2026, 5:08pm UTC](https://shapeworks.discourse.group/t/optimization-parameters-tuning/393/1 "2026-02-20T17:08:44Z")

</div>

I am trying to optimize this shape model, however the reconstruction is bad. The mean shape (with and without the particles) as well as some samples after reconstruction and optimization parameters are attached below. Would appreciate clues about which parameters I should focus on.

 ![image](https://global.discourse-cdn.com/free1/uploads/shapeworks/original/1X/3a4c7c3d8aff273aa3c94428e1231312fae59267.jpeg)

 ![image](https://global.discourse-cdn.com/free1/uploads/shapeworks/original/1X/2d42dbf9160d31c011c88d814e27faad88e16092.jpeg)

 ![image](https://global.discourse-cdn.com/free1/uploads/shapeworks/original/1X/1080807d8cbea4e569b9fde6ad1ddcdc0ea9c1df.png)

 ![image](https://global.discourse-cdn.com/free1/uploads/shapeworks/original/1X/0bb08b76763cf170565547f772ba18882988be49.png)

---

<div class="post-metadata">

**Author:** ![amorris](https://avatars.discourse-cdn.com/v4/letter/a/73ab20/32.png) [@amorris](https://shapeworks.discourse.group/u/amorris)\
**Post date:** [February 20, 2026, 7:51pm UTC](https://shapeworks.discourse.group/t/optimization-parameters-tuning/393/2 "2026-02-20T19:51:21Z")

</div>

The first thing I would suggest is to increase the normals strength. 0.5 is pretty low and they will not contribute much. Try perhaps 5 or 10.

---

<div class="post-metadata">

**Author:** ![blah4](https://avatars.discourse-cdn.com/v4/letter/b/7ea924/32.png) [@blah4](https://shapeworks.discourse.group/u/blah4)\
**Post date:** [February 23, 2026, 4:20pm UTC](https://shapeworks.discourse.group/t/optimization-parameters-tuning/393/3 "2026-02-23T16:20:21Z")

</div>

Hey, so I tried increasing the normals to 5, I see some improvement but the Mean Shape still does not look like a real vertebra.

 ![image](https://global.discourse-cdn.com/free1/uploads/shapeworks/original/1X/dfe80f03b0c99c09df238b9dec49d484e06e7f32.png)

 ![image](https://global.discourse-cdn.com/free1/uploads/shapeworks/original/1X/9760a64bead7845ca8c847fc7790a77c4df1c11b.jpeg)

 ![image](https://global.discourse-cdn.com/free1/uploads/shapeworks/original/1X/dab592859075093d26f7eb438148480ebab1f1a5.png)

---

<div class="post-metadata">

**Author:** ![amorris](https://avatars.discourse-cdn.com/v4/letter/a/73ab20/32.png) [@amorris](https://shapeworks.discourse.group/u/amorris)\
**Post date:** [February 25, 2026, 5:16pm UTC](https://shapeworks.discourse.group/t/optimization-parameters-tuning/393/4 "2026-02-25T17:16:30Z")

</div>

Is your main concern the thin parts of the reconstruction? You may just need more particles.