search

YuanSun-XJTU / iComMa

iComMa: Inverting 3D Gaussian Splatting for Camera Pose Estimation via Comparing and Matching
87 stars 4 forks source link

Troubleshooting Camera Pose Estimation: Analyzing Supervision Loss Design and Gradient Issues in Optimization #13

Open YufengJin opened 4 months ago

YufengJin commented 4 months ago

Dear,

I am also encountering unusual results in camera pose estimation. I replaced the differential rasterizer in the original GS work and pre-trained on the Blender dataset for 30,000 iterations, after which I saved the gaussians. Subsequently, I ceased the optimization of the gaussian model, focusing solely on adjusting the camera position. I introduced an error margin of +/- 0.2 meters and +/- 5 degrees to the camera, yet the final camera position merely fluctuates within a small range and fails to converge to the correct location. I utilized both RGB and depth as loss metrics. I am perplexed as to whether the issue lies in the design of the supervision loss or the gradient of the camera delta pose. Your insights would be highly appreciated.

whu-lyh commented 1 week ago

Hi do you solved this? I also met the same problems