Fix rendering energy loss in RenderEngineVtk

[SeanCurtis-TRI]

Fix rendering energy loss in RenderEngineVtk

Energy loss is due to three issues.

1. Drake introduced an erroneous patch which bled energy badly.
2. Set the skybox to use gamma correction, even for hdr images.
3. Introduce a new patch to change the VTK brdf model from a single-scatter
   approximation to a multi-scatter approximation.

This will have an immediate change on rendered outputs -- images will be brighter.

Based on brightness changes:

1. The lighting configuration tutorial has been updated accordingly.
2. The unit tests for RenderEngineVtk have had reference values (subject to PBR) brightened accordingly.

Current camera configurations will need to be updated, especially with respect to their exposure values and lighting levels. Renderings for which high exposure values seemed necessary will now seem oversaturated. Furthermore, with more energy coming from the environment map, lights may need to be brighter to have a comparable effect as before.

Resolves #23772

---

The following will let you see the effect:

1. Download an environment map (Drake doesn't store big expensive maps in its repo).

curl -o test.hdr https://dl.polyhaven.org/file/ph-assets/HDRIs/hdr/2k/thatch_chapel_2k.hdr

2. Use meshcat to exercise RenderEngineVtk:

bazel run //tools:model_visualizer_private  -- --show_rgbd_sensor --no_lights \
package://drake_models/iiwa_description/sdf/iiwa7_with_box_collision.sdf --environment_map test.hdr

When we run against master, we get the following:

This PR gives us:

In addition to the image just generally being brighter, the skybox with the environment map handles the dynamic lighting range better; details are more discernible through the window.

---

This change is 

[SeanCurtis-TRI]

+(release notes: fix) +a:@jwnimmer-tri for feature review, please.

[jwnimmer-tri]

@jwnimmer-tri reviewed 4 of 4 files at r1, all commit messages.
Reviewable status: 2 unresolved discussions, needs at least two assigned reviewers

---

a discussion (no related file):
Some unit test(s?) are complaining about the environment map.

---

tutorials/configuring_rendering_lighting.ipynb line 2645 at r1 (raw file):

    "    \"shadow_map_size\": 1024,\n",
    "    # With additional lights, we reduce exposure.\n",
    "    \"exposure\": 1, # more lights --> reduced exposure.\n",

BTW Doesn't this EOL comment duplicate the line immediately above?

[SeanCurtis-TRI]

Previously, jwnimmer-tri (Jeremy Nimmer) wrote…

Some unit test(s?) are complaining about the environment map.

Yeah...unit tests should be complaining. My oversight.

I've got a replacement patch (see #23772). I'm going to refresh this PR to not just remove the old, bad patch, but to include the new good patch.

Would you want to see that in a separate PR?

[jwnimmer-tri]

Reviewable status: 2 unresolved discussions, needs at least two assigned reviewers

---

a discussion (no related file):

Previously, SeanCurtis-TRI (Sean Curtis) wrote…

Yeah...unit tests should be complaining. My oversight.

I've got a replacement patch (see #23772). I'm going to refresh this PR to not just remove the old, bad patch, but to include the new good patch.

Would you want to see that in a separate PR?

I'm OK with any approach.

[SeanCurtis-TRI]

Reviewable status: 1 unresolved discussion, needs at least two assigned reviewers

---

a discussion (no related file):

Previously, SeanCurtis-TRI (Sean Curtis) wrote…

Yeah...unit tests should be complaining. My oversight.

I've got a replacement patch (see #23772). I'm going to refresh this PR to not just remove the old, bad patch, but to include the new good patch.

Would you want to see that in a separate PR?

I'll push it here and you can choose if you'd like it reverted. I like them both together because the two together solve the problem.

If you want to run the furnace test locally:

1. Download the following files:

curl -o material_quadrant.bin https://raw.githubusercontent.com/SeanCurtis-TRI/drake/refs/heads/PR_multiscatter_with_furnace/geometry/test/material_quadrant.bin
curl -o material_quadrant.gltf https://raw.githubusercontent.com/SeanCurtis-TRI/drake/refs/heads/PR_multiscatter_with_furnace/geometry/test/material_quadrant.gltf
curl -o furnace.hdr https://raw.githubusercontent.com/SeanCurtis-TRI/drake/refs/heads/PR_multiscatter_with_furnace/geometry/test/furnace.hdr

2. Run the visualizer

bazel run //tools:model_visualizer_private  -- --show_rgbd_sensor /absolute/path/to/the/files/you/downloaded/material_quadrant.gltf  --environment_map furnace.hdr --no_light

Running that on master and this PR will illustrate the lighting change (in ways a robot in a noisy environment can't).

[jwnimmer-tri]

Re-

@jwnimmer-tri reviewed 5 of 5 files at r2, all commit messages.
Reviewable status: 1 unresolved discussion, needs at least two assigned reviewers

---

a discussion (no related file):

Previously, SeanCurtis-TRI (Sean Curtis) wrote…

I'll push it here and you can choose if you'd like it reverted. I like them both together because the two together solve the problem.

If you want to run the furnace test locally:

1. Download the following files:

curl -o material_quadrant.bin https://raw.githubusercontent.com/SeanCurtis-TRI/drake/refs/heads/PR_multiscatter_with_furnace/geometry/test/material_quadrant.bin
curl -o material_quadrant.gltf https://raw.githubusercontent.com/SeanCurtis-TRI/drake/refs/heads/PR_multiscatter_with_furnace/geometry/test/material_quadrant.gltf
curl -o furnace.hdr https://raw.githubusercontent.com/SeanCurtis-TRI/drake/refs/heads/PR_multiscatter_with_furnace/geometry/test/furnace.hdr

2. Run the visualizer

bazel run //tools:model_visualizer_private  -- --show_rgbd_sensor /absolute/path/to/the/files/you/downloaded/material_quadrant.gltf  --environment_map furnace.hdr --no_light

Running that on master and this PR will illustrate the lighting change (in ways a robot in a noisy environment can't).

I agree together is best.

In any case, the CI is fixed so I'll close this thread.

---

tools/workspace/vtk_internal/repository.bzl line 192 at r3 (raw file):

            #   will be upstreamed into VTK itself, so they should be the first
            #   changes applied to reduce merge conflict churn.
            # - Patch file names should begin with the name of the module being

BTW Maybe rename the patch file to match this convention?

(Also I wonder if "add" is the right verb for it. Aren't we replacing single-scatter with multi-?)

Code quote:

Patch file names should begin with the name of the module

[SeanCurtis-TRI]

BTW This is the PR I had in mind for using the release notes: announce tag (#23812).

+a:@sammy-tri for platform review, please.

Reviewable status: LGTM missing from assignee sammy-tri(platform)

[jwnimmer-tri]

@jwnimmer-tri reviewed 2 of 2 files at r4, all commit messages.
Reviewable status: LGTM missing from assignee sammy-tri(platform)

[sammy-tri]

@sammy-tri reviewed 2 of 4 files at r1, 3 of 5 files at r2, 2 of 2 files at r4, all commit messages.
Reviewable status: complete! all discussions resolved, LGTM from assignees sammy-tri(platform),jwnimmer-tri(platform)