SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2108.05312 · ICCV · 2021

Towards Interpretable Deep Networks for Monocular Depth Estimation

Wei-Chen Chiu, Yi-Hsuan Tsai, Guanbin Li, Zunzhi You

arXiv · PDF · Open in the Atlas

Code that ran

5 of 5 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (4 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 5 have no contract verdict; 0 failures of our instrument; 0 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2108.05312.json · badge: /api/ran/2108.05312.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
youzunzhi/InterpretableMDE community
Papers with Code archive
5 ran · 0 no recorded run
FunctionStatusWhere it lives
atrous_conv Ran
ran on a synthesized input · constructed an object; no result was computed
youzunzhi/InterpretableMDE/model/bts/modules.py
code served (permissive licence) · get_code("a7351e4a2a751444")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, atrous_conv: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    'in_channels': 3,
    'out_channels': 16,
    'dilation': 2,
    'apply_bn_first': True,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type atrous_conv — the run constructed an instance of its own class and did not call it; no shape is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
bts Ran
ran on a synthesized input · constructed an object; no result was computed
youzunzhi/InterpretableMDE/model/bts/modules.py
code served (permissive licence) · get_code("8a49b510e671008e")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, bts: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
import math

torch.manual_seed(0)

# Small stand-in classes and minimal setup
class DummyDataset:
    def __init__(self):
        self.name = "nyudepthv2"

# Create minimal feature extractor that outputs tensors with appropriate channel dimensions
class DummyFeatureExtractor(nn.Module):
    def __init__(self, feat_out_channels):
        super(DummyFeatureExtractor, self).__init__()
        self.feat_out_channels = feat_out_channels
        # Create dummy conv layers for each feature level
        self.convs = nn.ModuleList([
            nn.Conv2d(3, feat_out_channels[0], 3, 1, 1),
            nn.Conv2d(feat_out_channels[0], feat_out_channels[1], 3, 1, 1),
            nn.Conv2d(feat_out_channels[1], feat_out_channels[2], 3, 1, 1),
            nn.Conv2d(feat_out_channels[2], feat_out_channels[3], 3, 1, 1),
            nn.Conv2d(feat_out_channels[3], feat_out_channels[4], 3, 1, 1),
        ])
    
    def forward(self, x):
        features = []
        out = x
        for conv in self.convs:
            out = conv(out)
            features.append(out)
        return features

# Define arguments for bts.__init__
ARGS = {
    'feat_out_channels': [64, 128, 256, 512, 1024],  # Channel dimensions for each feature level
    'dataset': DummyDataset(),  # Dataset object with name attribute
    'max_depth': 10.0,  # Maximum depth value
    'num_features': 512,  # Number of features (default)
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type bts — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a bts output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
local_planar_guidance Ran
ran on a synthesized input
youzunzhi/InterpretableMDE/model/bts/modules.py
code served (permissive licence) · get_code("238cbbeee2487336")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 3, dtype family float; no output shape is stated. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

class local_planar_guidance(nn.Module):

    def __init__(self, upratio):
        super(local_planar_guidance, self).__init__()
        self.upratio = upratio
        self.u = torch.arange(self.upratio).reshape([1, 1, self.upratio]).float()
        self.v = torch.arange(int(self.upratio)).reshape([1, self.upratio, 1]).float()
        self.upratio = float(upratio)

    def forward(self, plane_eq, focal):
        plane_eq_expanded = torch.repeat_interleave(plane_eq, int(self.upratio), 2)
        plane_eq_expanded = torch.repeat_interleave(plane_eq_expanded, int(self.upratio), 3)
        n1 = plane_eq_expanded[:, 0, :, :]
        n2 = plane_eq_expanded[:, 1, :, :]
        n3 = plane_eq_expanded[:, 2, :, :]
        n4 = plane_eq_expanded[:, 3, :, :]
        u = self.u.repeat(plane_eq.size(0), plane_eq.size(2) * int(self.upratio), plane_eq.size(3))
        if torch.cuda.is_available():
            u = u.cuda()
        u = (u - (self.upratio - 1) * 0.5) / self.upratio
        v = self.v.repeat(plane_eq.size(0), plane_eq.size(2), plane_eq.size(3) * int(self.upratio))
        if torch.cuda.is_available():
            v = v.cuda()
        v = (v - (self.upratio - 1) * 0.5) / self.upratio
        return n4 / (n1 * u + n2 * v + n3)


INIT_ARGS = {
    'upratio': 3
}

ARGS = {
    'plane_eq': torch.randn(2, 4, 5, 6),
    'focal': torch.randn(2, 1)
}
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 3, dtype family float — no output shape is stated: observe_signature.py observed this class with its constructor sizes canonicalised where it could, which the node does not record, so its shape is not stated as this setup block's
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
reduction_1x1 Ran
ran on a synthesized input · constructed an object; no result was computed
youzunzhi/InterpretableMDE/model/bts/modules.py
code served (permissive licence) · get_code("7a747dbea815af1b")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, reduction_1x1: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
import math

torch.manual_seed(0)

ARGS = {
    'num_in_filters': 16,
    'num_out_filters': 8,
    'max_depth': 10.0,
    'is_final': False
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type reduction_1x1 — the run constructed an instance of its own class and did not call it; no shape is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
upconv Ran
ran on a synthesized input · constructed an object; no result was computed
youzunzhi/InterpretableMDE/model/bts/modules.py
code served (permissive licence) · get_code("49763904bc51bfe9")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, upconv: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
import torch.nn.functional as torch_nn_func

torch.manual_seed(0)

# Create a sample input tensor
# The upconv module expects input with shape (batch, channels, height, width)
# Let's use batch=2, in_channels=4, height=8, width=8
sample_input = torch.randn(2, 4, 8, 8)

ARGS = {
    'in_channels': 4,
    'out_channels': 3,
    'ratio': 2
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type upconv — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a upconv output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 5 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Deep networks for Monocular Depth Estimation (MDE) have achieved promising performance recently and it is of great importance to further understand the interpretability of these networks. Existing methods attempt to provide posthoc explanations by investigating visual cues, which may not explore the internal representations learned by deep networks. In this paper, we find that some hidden units of the network are selective to certain ranges of depth, and thus such behavior can be served as a way to interpret the internal representations. Based on our observations, we quantify the interpretability of a deep MDE network by the depth selectivity of its hidden units. Moreover, we then propose a method to train interpretable MDE deep networks without changing their original architectures, by assigning a depth range for each unit to select. Experimental results demonstrate that our method is able to enhance the interpretability of deep MDE networks by largely improving the depth selectivity of their units, while not harming or even improving the depth estimation accuracy. We further provide comprehensive analysis to show the reliability of selective units, the applicability of our method on different layers, models, and datasets, and a demonstration on analysis of model error. Source code and models are available at https:// github.com/youzunzhi/InterpretableMDE.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2108.05312")
get_code_for_paper("2108.05312")
have("2108.05312")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2108.05312.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2108.05312.svg)](https://syntology.ai/paper/2108.05312)

Connect an agent — have() is free.