SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2108.10552 · 2021

E-RAFT: Dense Optical Flow from Event Cameras

arXiv · PDF · Open in the Atlas

Code that ran

10 of 12 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 1 honoured the contract we drafted, 0 violated it and 8 have no contract verdict; 1 failure of our instrument; 2 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/2108.10552.json · badge: /api/ran/2108.10552.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
uzh-rpg/e-raft canonical (classifier label)
classifier basis: census
10 ran · 2 no recorded run
FunctionStatusWhere it lives
coords_grid Ran
ran, honoured the contract we drafted
uzh-rpg/e-raft/model/utils.py
pointer only (licence: MIT); the stored copy get_code serves, from uci-cbcl/RP-Net, is pointer only (licence: NONE) · get_code("7c8c4c5cb4271dfa")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from uci-cbcl/RP-Net, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

flow_16bit_to_float Ran
ran on a synthesized input
uzh-rpg/e-raft/utils/dsec_utils.py
code served (permissive licence) · get_code("efe587c026e07812")
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 tuple. 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 numpy as np

# Create a small valid 16-bit flow array
# Shape: (height=4, width=3, channels=3)
# Channel 0: x-component of flow (uint16)
# Channel 1: y-component of flow (uint16)
# Channel 2: validity mask (1 for valid, 0 for invalid)

h, w = 4, 3
flow_16bit = np.zeros((h, w, 3), dtype=np.uint16)

# Set some valid pixels (channel 2 = 1)
flow_16bit[0, 0, 2] = 1
flow_16bit[0, 1, 2] = 1
flow_16bit[1, 0, 2] = 1
flow_16bit[2, 2, 2] = 1

# Set flow values for valid pixels
# Values should be around 2^15 (32768) so that after subtracting 2^15 they become reasonable
flow_16bit[0, 0, 0] = 32768 + 128  # x flow = 1.0
flow_16bit[0, 0, 1] = 32768 - 128  # y flow = -1.0

flow_16bit[0, 1, 0] = 32768 + 256  # x flow = 2.0
flow_16bit[0, 1, 1] = 32768        # y flow = 0.0

flow_16bit[1, 0, 0] = 32768        # x flow = 0.0
flow_16bit[1, 0, 1] = 32768 + 256  # y flow = 2.0

flow_16bit[2, 2, 0] = 32768 - 256  # x flow = -2.0
flow_16bit[2, 2, 1] = 32768 + 512  # y flow = 4.0

# Ensure invalid pixels have channel 2 = 0
flow_16bit[~(flow_16bit[..., 2] == 1), 2] = 0

ARGS = {
    "flow_16bit": flow_16bit
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none 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
forward_interpolate_pytorch Ran
ran on a synthesized input
uzh-rpg/e-raft/utils/image_utils.py
code served (permissive licence) · get_code("6fc5ee098dfefa3b")
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 Tensor, ndim 4, dtype family float, shape (2, 2, 4, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape 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

torch.manual_seed(0)

# Create a small flow tensor with shape (batch=2, channels=2, height=4, width=3)
# The function expects flow_in with optical flow data (dx, dy)
flow_in = torch.randn(2, 2, 4, 3)

ARGS = {
    "flow_in": flow_in
}
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 Tensor, ndim 4, dtype family float, shape (2, 2, 4, 3) — shape (2, 2, 4, 3) as a later run recorded it
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: none by the harness; any seed is in the setup block
get_image Ran
ran on a synthesized input
uzh-rpg/e-raft/loader/utils.py
code served (permissive licence) · get_code("ca480d2713282e55")
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 ndarray, ndim 2, dtype family int, shape (2, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape 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 numpy
from PIL import Image
import io

# Create a small test image in memory
img_array = numpy.uint8([[100, 150, 200], [50, 75, 125]])
img = Image.fromarray(img_array)

# Save to bytes buffer
buffer = io.BytesIO()
img.save(buffer, format='PNG')
buffer.seek(0)

# Create a temporary file-like object that can be opened by PIL
import tempfile
import os

# Write test image to a temporary file
temp_file = tempfile.NamedTemporaryFile(suffix='.png', delete=False)
temp_file.write(buffer.getvalue())
temp_file.close()

ARGS = {
    'image_path': temp_file.name
}

# Cleanup will happen after function execution
# The temporary file path is valid for the function call
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family int, shape (2, 3) — shape (2, 3) as a later run recorded it
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: none by the harness; any seed is in the setup block
get_ts Ran
ran on a synthesized input
uzh-rpg/e-raft/loader/utils.py
code served (permissive licence) · get_code("bcff40acfef90b26")
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 int, ndim 0, dtype family int, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape 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 tempfile
import os

# Create a temporary file with test data
temp_file = tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt')
temp_file.write("42\n")
temp_file.write("3.14\n")
temp_file.write("100\n")
temp_file.close()

ARGS = {
    'path': temp_file.name,
    'i': 0,
    'type': 'int'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type int, ndim 0, dtype family int, shape () — shape () as a later run recorded it
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: none by the harness; any seed is in the setup block
get_values_from_key Ran
ran on a synthesized input
uzh-rpg/e-raft/utils/helper_functions.py
code served (permissive licence) · get_code("d81d5e29bb21898b")
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 list, ndim 1, dtype family float; no output shape is stated. The image it ran in was 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 numpy as np

torch.manual_seed(0)

# The function iterates through input_list and accesses i[key] for each element
# So input_list should be a list of dict-like objects (dicts or objects with __getitem__)
# and key should be a valid key to access

input_list = [
    {"name": "Alice", "age": 30},
    {"name": "Bob", "age": 25},
    {"name": "Charlie", "age": 35}
]

key = "name"

ARGS = {
    "input_list": input_list,
    "key": key
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
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: none by the harness; any seed is in the setup block
grid_sample_values Ran
ran on a synthesized input
uzh-rpg/e-raft/utils/image_utils.py
code served (permissive licence) · get_code("24dcd89dbfba92ed")
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 tuple, ndim 4, dtype family float; no output shape is stated. The image it ran in was 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

torch.manual_seed(0)

# The function expects:
# - input: a tensor with shape (3, N) where N is number of points
#   - input[0, :] are x coordinates
#   - input[1, :] are y coordinates
#   - input[2, :] are z values
# - height: int, height of grid
# - width: int, width of grid

# Create a small input tensor with 3 rows (x, y, z) and 4 points
input = torch.tensor([
    [0.5, 1.2, 2.7, 3.1],  # x coordinates
    [0.3, 1.5, 2.2, 0.8],  # y coordinates
    [10.0, 20.0, 15.0, 25.0]  # z values
], dtype=torch.float32)

height = 5
width = 6

ARGS = {
    "input": input,
    "height": height,
    "width": width,
}
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 tuple, ndim 4, dtype family float — a tuple output has no array shape of its own; none is stated
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: none by the harness; any seed is in the setup block
move_dict_to_cuda Ran
ran on a synthesized input
uzh-rpg/e-raft/utils/helper_functions.py
code served (permissive licence) · get_code("06a201456bf102b9")
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 dict. 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

torch.manual_seed(0)

# Create a nested dictionary of tensors
dictionary_of_tensors = {
    'tensor_a': torch.randn(2, 4),
    'tensor_b': torch.randn(2, 3),
    'nested': {
        'tensor_c': torch.randn(2, 5),
        'tensor_d': torch.randn(2, 2)
    }
}

# GPU specification - using 'cpu' since CUDA is not available
gpu = 'cpu'

ARGS = {
    'dictionary_of_tensors': dictionary_of_tensors,
    'gpu': gpu
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
move_list_to_cuda Ran
ran on a synthesized input
uzh-rpg/e-raft/utils/helper_functions.py
code served (permissive licence) · get_code("13d7e5935de17ce1")
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 list, ndim 1, dtype family float; no output shape is stated. The image it ran in was 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

torch.manual_seed(0)

# Construct a list of dictionaries containing tensors
list_of_dicts = [
    {
        "tensor_a": torch.randn(2, 4),
        "tensor_b": torch.randn(2, 3),
        "nested": {
            "tensor_c": torch.randn(2, 5)
        }
    },
    {
        "tensor_d": torch.randn(2, 4),
        "nested": {
            "tensor_e": torch.randn(2, 6),
            "deeper": {
                "tensor_f": torch.randn(2, 7)
            }
        }
    }
]

# GPU device specification (cpu for testing)
gpu = "cpu"

ARGS = {
    "list_of_dicts": list_of_dicts,
    "gpu": gpu
}
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 list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
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: none by the harness; any seed is in the setup block
upflow8 Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
uzh-rpg/e-raft/model/utils.py
pointer only (licence: MIT); the stored copy get_code serves, from plusgood-steven/id-blau, is pointer only (licence: NONE) · get_code("561d1e8d3102a282")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from plusgood-steven/id-blau, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

bilinear_sampler Not yet run
unverified (harvested, no recorded run)
uzh-rpg/e-raft/model/utils.py
code served (permissive licence) · get_code("1ac8cce613d98769")
How it ran

no invocation recorded.

get_events Not yet run
unverified (harvested, no recorded run)
uzh-rpg/e-raft/loader/utils.py
code served (permissive licence) · get_code("ba22c483efb8b757")
How it ran

no invocation recorded.

Also present in other papers' code

Of 12 samples lifted from this paper's linked repositories, 12 carry a function identity and 0 do not. Of the 12 distinct functions with an identity, 3 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 2 the same month, 3 later) and 9 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.

FunctionAlso present inEarliest-datedOthers named
bilinear_sampler 18 other papers
1 dated earlier, 1 the same month, 16 later · 7 only as an assembly brick
2003.12039 (2020-03)
RAFT: Recurrent All-Pairs Field Transforms for Optical Flow
2108.00622 (2021-08)
Recurrent Mask Refinement for Few-Shot Medical Image Segmentation
2109.15321 (2021-09)
Sensor-Guided Optical Flow
and 15 more not named here
coords_grid 6 other papers
1 the same month, 5 later · 4 only as an assembly brick
2108.00622 (2021-08)
Recurrent Mask Refinement for Few-Shot Medical Image Segmentation
2109.15321 (2021-09)
Sensor-Guided Optical Flow
2203.11335 (2022-03)
Global Matching with Overlapping Attention for Optical Flow Estimation
and 3 more not named here
upflow8 5 other papers
1 dated earlier, 4 later · 4 only as an assembly brick
2003.12039 (2020-03)
RAFT: Recurrent All-Pairs Field Transforms for Optical Flow
2202.03857 (2022-02)
Learning Optical Flow with Adaptive Graph Reasoning
2203.11335 (2022-03)
Global Matching with Overlapping Attention for Optical Flow Estimation
and 2 more not named here

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

We propose to incorporate feature correlation and sequential processing into dense optical flow estimation from event cameras. Modern frame-based optical flow methods heavily rely on matching costs computed from feature correlation. In contrast, there exists no optical flow method for event cameras that explicitly computes matching costs. Instead, learning-based approaches using events usually resort to the U-Net architecture to estimate optical flow sparsely. Our key finding is that the introduction of correlation features significantly improves results compared to previous methods that solely rely on convolution layers. Compared to the state-of-the-art, our proposed approach computes dense optical flow and reduces the end-point error by 23% on MVSEC. Furthermore, we show that all existing optical flow methods developed so far for event cameras have been evaluated on datasets with very small displacement fields with a maximum flow magnitude of 10 pixels. Based on this observation, we introduce a new real-world dataset that exhibits displacement fields with magnitudes up to 210 pixels and 3 times higher camera resolution. Our proposed approach reduces the end-point error on this dataset by 66%.

For agents

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

get_harvested_code_for_paper("2108.10552")
get_code_for_paper("2108.10552")
have("2108.10552")

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

curl https://syntology.ai/api/ran/2108.10552.json

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

[![Syntology run record](https://syntology.ai/api/ran/2108.10552.svg)](https://syntology.ai/paper/2108.10552)

Connect an agent — have() is free.