Comments (4)
Sorry, now I understand: the gradient is backpropagated to depth and pose through the calculated weights wa, wb, wc, wd
, not through the indices idx_a, idx_b, idx_c, idx_d
, i.e., through the interpolation. Of course, for each sampled pixel, the net decides "where to move" in the grid only looking at the four neighbouring pixels used in the interpolation, but not beyond those pixels. So:
Am I wrong if I think that the optimization works because the baseline between consecutive frames is small? I mean, if the baseline were greater, the optimization wouldn't work, right?
Cheers!
from sfmlearner.
Oh, why do you think the gradients cannot backprop through tf.gather
? FYI, I copied the _interpolate
method from the official tensorflow repo: https://github.com/tensorflow/models/blob/master/transformer/spatial_transformer.py.
from sfmlearner.
You are right that the gradients for depth and pose are backproped through the w's, and the gradient can only be derived from the four neighboring pixels used in the interpolation, which is why we use a multi-scale loss and depth smoothness loss in the objective to overcome the gradient locality (see Sec. 3.4 of the paper). With multi-scale and smoothness, the system is able to deal with reasonably large baselines.
from sfmlearner.
Oh, I didn't take that into account. Thank you very much for the explanation and for sharing your code!
Cheers!
from sfmlearner.
Related Issues (20)
- ONNX Export HOT 1
- Static_frames.txt
- Question of depth output
- ValueError: cannot reshape array of size 1 into shape (3,3) HOT 1
- The structure of datasets HOT 1
- when I use the pre-trained pose model to predict, the output is same whatever input are HOT 3
- when I use the pre-trained pose model to predict, the output is same whatever inputs are.
- The website download the pretrained model got a 404 error HOT 2
- About posenet
- understand the pose estimation values (Absolute Trajectory Error (ATE))
- How to get ATE values as in Table 3 in your paper
- Is the dispnet output a depth map?
- Question about paper (projected coordinates formula) HOT 2
- Pre-processing NYUv2 dataset
- question about dump_xyz in evaluate_pose.py
- How to get an output of the depth or pose of a pixel away from the camera?
- Question about diff image in tensorboard HOT 1
- poseNet model reasoning speed problem
- Why pose GT you provided differ from KITTI site?
- About evaluation on the Make3D dataset?
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from sfmlearner.