SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2203.13296 · ECCV · 2022

RayTran: 3D pose estimation and shape reconstruction of multiple objects from videos with ray-traced transformers

M Tyszkiewicz, K.-K Maninis, S Popov, V Ferrari

arXiv · PDF · Open in the Atlas

Code that ran

4 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 4 have no contract verdict; 0 failures of our instrument; 1 no recorded run), measured 2026-11-22; 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/2203.13296.json · badge: /api/ran/2203.13296.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
likojack/ODAM community
Papers with Code archive
4 ran · 1 no recorded run
FunctionStatusWhere it lives
TransformerDecoder Ran
ran on a synthesized input · constructed an object; no result was computed
likojack/ODAM/src/models/transformer.py
code served (permissive licence) · get_code("32c05f7bbeafcdaf")
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, TransformerDecoder: 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
from torch import Tensor
import copy

torch.manual_seed(0)

# Create a minimal decoder layer that matches the expected interface
class MinimalDecoderLayer(nn.Module):
    def __init__(self, d_model=4, nhead=2):
        super().__init__()
        self.self_attn = nn.MultiheadAttention(d_model, nhead, batch_first=True)
        self.cross_attn = nn.MultiheadAttention(d_model, nhead, batch_first=True)
        self.linear1 = nn.Linear(d_model, 8)
        self.linear2 = nn.Linear(8, d_model)
        self.norm1 = nn.LayerNorm(d_model)
        self.norm2 = nn.LayerNorm(d_model)
        self.norm3 = nn.LayerNorm(d_model)
        
    def forward(self, tgt, memory, tgt_mask=None, memory_mask=None, 
                tgt_key_padding_mask=None, memory_key_padding_mask=None, 
                pos=None, query_pos=None):
        # Self attention
        tgt2, _ = self.self_attn(tgt, tgt, tgt, attn_mask=tgt_mask, 
                                 key_padding_mask=tgt_key_padding_mask)
        tgt = tgt + tgt2
        tgt = self.norm1(tgt)
        
        # Cross attention
        tgt2, _ = self.cross_attn(tgt, memory, memory, attn_mask=memory_mask,
                                  key_padding_mask=memory_key_padding_mask)
        tgt = tgt + tgt2
        tgt = self.norm2(tgt)
        
        # FFN
        tgt2 = self.linear2(torch.relu(self.linear1(tgt)))
        tgt = tgt + tgt2
        tgt = self.norm3(tgt)
        
        return tgt

# Create decoder layer
decoder_layer = MinimalDecoderLayer(d_model=4, nhead=2)

# Create normalization layer
norm_layer = nn.LayerNorm(4)

# Create sample tensors
batch_size = 2
seq_len_tgt = 3
seq_len_mem = 5
d_model = 4

tgt = torch.randn(batch_size, seq_len_tgt, d_model)
memory = torch.randn(batch_size, seq_len_mem, d_model)

ARGS = {
    'decoder_layer': decoder_layer,
    'num_layers': 2,
    'norm': norm_layer,
    'return_intermediate': 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 TransformerDecoder — 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 TransformerDecoder 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
TransformerDecoderLayer Ran
ran on a synthesized input · constructed an object; no result was computed
likojack/ODAM/src/models/transformer.py
code served (permissive licence) · get_code("df37859c9b2a3024")
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, TransformerDecoderLayer: 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)

# Define the arguments for TransformerDecoderLayer
ARGS = {
    'd_model': 4,
    'nhead': 2,
    'dim_feedforward': 8,
    'dropout': 0.1,
    'activation': 'relu',
    'normalize_before': 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 TransformerDecoderLayer — 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 TransformerDecoderLayer 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
TransformerEncoder Ran
ran on a synthesized input · constructed an object; no result was computed
likojack/ODAM/src/models/transformer.py
code served (permissive licence) · get_code("ec86d33233d305b6")
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, TransformerEncoder: 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
from typing import Optional

torch.manual_seed(0)

# Define a minimal encoder layer that matches what TransformerEncoder expects
class MinimalEncoderLayer(nn.Module):
    def __init__(self, d_model=4):
        super().__init__()
        self.linear = nn.Linear(d_model, d_model)
    
    def forward(self, src, src_mask=None, src_key_padding_mask=None, pos=None):
        # The layer receives src and optional mask/pos arguments
        # Just apply a simple transformation
        return self.linear(src)

# Create the encoder layer instance
encoder_layer = MinimalEncoderLayer(d_model=4)

# Number of layers to stack
num_layers = 2

# Optional normalization layer
norm = nn.LayerNorm(4)

ARGS = {
    "encoder_layer": encoder_layer,
    "num_layers": num_layers,
    "norm": norm
}
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 TransformerEncoder — 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 TransformerEncoder 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
TransformerEncoderLayer Ran
ran on a synthesized input · constructed an object; no result was computed
likojack/ODAM/src/models/transformer.py
code served (permissive licence) · get_code("110d117a309e1201")
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, TransformerEncoderLayer: 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
from typing import Optional

torch.manual_seed(0)

# Define ARGS dict with all parameters for TransformerEncoderLayer
ARGS = {
    'd_model': 4,
    'nhead': 2,
    'dim_feedforward': 8,
    'dropout': 0.1,
    'activation': 'relu',
    'normalize_before': 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 TransformerEncoderLayer — 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 TransformerEncoderLayer 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
Transformer Not yet run
unverified (harvested, no recorded run)
likojack/ODAM/src/models/transformer.py
code served (permissive licence) · get_code("1c4022d68db2655a")
How it ran

no invocation recorded.

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, 4 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 1 the same month, 3 later) and 1 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
TransformerEncoder
brick of this paper's assembly: model
16 other papers
7 dated earlier, 2 the same month, 7 later · 7 only as an assembly brick
2005.12872 (2020-05)
End-to-End Object Detection with Transformers
2103.11468 (2021-03)
Learning Multi-Scene Absolute Pose Regression with Transformers
2103.12236 (2021-03)
Instance-level Image Retrieval using Reranking Transformers
and 13 more not named here
TransformerEncoderLayer
brick of this paper's assembly: model
16 other papers
5 dated earlier, 11 later · 8 only as an assembly brick
2005.12872 (2020-05)
End-to-End Object Detection with Transformers
2103.11468 (2021-03)
Learning Multi-Scene Absolute Pose Regression with Transformers
2104.02439 (2021-04)
Few-Shot Transformation of Common Actions into Time and Space
and 13 more not named here
TransformerDecoderLayer
brick of this paper's assembly: model
6 other papers
4 dated earlier, 2 later · 3 only as an assembly brick
2005.12872 (2020-05)
End-to-End Object Detection with Transformers
2103.11468 (2021-03)
Learning Multi-Scene Absolute Pose Regression with Transformers
2104.03964 (2021-04)
Handwriting Transformers
and 3 more not named here
Transformer
brick of this paper's assembly: entry
1 other paper
1 dated earlier
2108.07084 (2021-08)
Learning Canonical View Representation for 3D Shape Recognition with Arbitrary Views

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 a transformer-based neural network architecture for multi-object 3D reconstruction from RGB videos. It relies on two alternative ways to represent its knowledge: as a global 3D grid of features and an array of view-specific 2D grids. We progressively exchange information between the two with a dedicated bidirectional attention mechanism. We exploit knowledge about the image formation process to significantly sparsify the attention weight matrix, making our architecture feasible on current hardware, both in terms of memory and computation. We attach a DETR-style head [9] on top of the 3D feature grid in order to detect the objects in the scene and to predict their 3D pose and 3D shape. Compared to previous methods, our architecture is single stage, end-to-end trainable, and it can reason holistically about a scene from multiple video frames without needing a brittle tracking step. We evaluate our method on the challenging Scan2CAD dataset [3], where we outperform (1) state-of-the-art methods [39,34,35,15] for 3D object pose estimation from RGB videos; and (2) a strong alternative method combining with RGB-D CAD alignment [4].

For agents

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

get_harvested_code_for_paper("2203.13296")
get_code_for_paper("2203.13296")
have("2203.13296")

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

curl https://syntology.ai/api/ran/2203.13296.json

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

[![Syntology run record](https://syntology.ai/api/ran/2203.13296.svg)](https://syntology.ai/paper/2203.13296)

Connect an agent — have() is free.