Oliver Batchelor

Results 35 issues of Oliver Batchelor

Is it possible you could you share some of the trained models you have created?

Is it possible to configure OpenSfM to use a local coordinate system, rather than a GPS lat/long etc.? For example using poses from a robotic gantry or output from pipeline...

enhancement
question

EventWriterT recently changed: https://github.com/reflex-frp/reflex/commit/ca50bb4cf1e85e1c980b3ddad0e1b0c483542349#diff-ca34969f303705bc74cc027f0eedb19f Which after some digging I think is a more correct behaviour (even though it broke my code!). Except that RequesterT still has the old behaviour too,...

bug
High Severity

I had this kicking around, so I figured I might as well add a couple of tests and submit it for completeness. It seems useful for writing containers. One very...

enhancement
opinionated
breaking change

This allows the Event part of the output Dynamic to also depend on the Behaviour part of the input Dynamic. It's a generalisation of the existing buildDynamic.. It seems like...

enhancement
question
Low Severity

The idea would be to use reflexes hold initialization queue to properly order Dynamics in a way which avoids the need to worry about sampling the initial value. As a...

enhancement
Low Severity

Currently there're different MonadHolds (HostFrame, PushM) which don't interoperate with them, would it be reasonable to have a hierarchy of different types, and what would it look like? HostFrame is...

enhancement
Medium Severity

#### Changes Fixes #1597 Some tweaks which allow various detectors to run in 16 bit or mixed precision mode. - Ununsed num_features parameters - pytest api changes (pytest.approx?) #### Type...

### Describe the bug ``` File "/local/miniconda3/envs/pytorch10/lib/python3.9/site-packages/kornia/feature/scale_space_detector.py", line 223, in forward responses, lafs = self.detect(img, self.num_features, mask) File "/local/miniconda3/envs/pytorch10/lib/python3.9/site-packages/kornia/feature/scale_space_detector.py", line 165, in detect responses_flatten = response_max.view(response_max.size(0), -1) # [B, N]...

bug :bug:
help wanted
module: feature

I've optimized feature extraction and feature matching to make better use of the GPU by reading inputs and writing outputs in threads. * Makes feature extractors more fully use the...